| |
Lopinavir/ritonavir monotherapy or plus zidovudine and lamivudine in antiretroviral-naive HIV-infected patients
|
| |
| |
AIDS:Volume 22(3)30 January 2008p 385-393
Delfraissy, Jean-Francoisa; Flandre, Philippeb; Delaugerre, Constancec; Ghosn, Jadea; Horban, Andrzejd; Girard, Pierre-Mariee; Norton, Michaelf; Rouzioux, Christinec; Taburet, Anne-Marieg; Cohen-Codar, Isabellef; Van, Philippe Ngof; Chauvin, Jean-Pierref
From the aAP-HP, Department of Internal Medicine and Infectious Diseases, Bicetre University Hospital, Le Kremlin-Bicetre, France
bINSERM U 720, Pierre and Marie Curie University, Paris, France
cAP-HP, Virology Department, Necker University Hospital, Paris, France
dDepartment of Infectious Diseases, Warsaw, Poland
eAP-HP, Department of Infectious Diseases, Saint-Antoine University Hospital, Paris, France
fAbbott Laboratories, Rungis, France
gAP-HP, Laboratory of Clinical Pharmacology, Bicetre University Hospital, Le Kremlin-Bicetre, France.
Abstract
Background: Guidelines for the use of antiretroviral agents for HIV-1 infection recommend combining at least three agents. The toxicity, cost, and complexity of such regimens warrant the search for other options.
Methods: MONARK is a prospective, open-label, randomized, 96-week trial comparing the safety and efficacy of lopinavir/ritonavir monotherapy with a standard lopinavir/ritonavir plus zidovudine and lamivudine regimen as an initial treatment regimen in HIV-infected patients with HIV-RNA levels less than 100 000 copies/ml. The primary endpoint was the proportion of patients with HIV-1-RNA levels below 400 copies/ml at week 24 and below 50 copies/ml at week 48.
Results: Eight-three and 53 patients were randomly assigned and exposed in the monotherapy and triple-drug groups, respectively. At week 48, by an intent-to-treat analysis, 53 of 83 patients (64%) in the monotherapy group and 40 of 53 patients (75%) in the triple-drug group achieved the primary endpoint (P = 0.19). The on-treatment analysis indicates that 80 and 95% of patients reached the primary endpoint in the monotherapy and triple-drug groups, respectively (P = 0.02). In the monotherapy arm, protease inhibitor-associated resistance mutations were seen in three of the 21 patients qualifying for genotypic resistance testing, with a modest impact on lopinavir susceptibility. None of the serious reported adverse events were considered to be related to study treatment.
Conclusion: Our results suggest that lopinavir/ritonavir monotherapy demonstrates lower rates of virological suppression when compared with lopinavir/ritonavir triple therapy and therefore should not be considered as a preferred treatment option for widespread use in antiretroviral-naive patients.
Discussion
This is the first randomized trial comparing LPV/r monotherapy as a first-line regimen with a recommended potent triple regimen of zidovudine, lamivudine and LPV/r [15]. In these antiretroviral-naive patients, using ITT-E analysis, 64% of patients randomly assigned reached the protocol-defined virological response in the LPV/r monotherapy arm compared with 75% in the LPV/r triple therapy arm. This difference was not statistically significant. The on-treatment analysis at week 48, however, shows a larger discrepancy between the two randomized groups with 80 and 98% of patients reaching the primary endpoint in the LPV/r monotherapy arm versus the LPV/r triple therapy arm. LPV/r monotherapy resulted in similar CD4 cell count increases as obtained with the standard triple drug combination. Patients on LPV/r monotherapy reported significantly fewer symptoms than those on triple combination.
In the LPV/r triple therapy arm, by ITT-E analysis, the percentage of patients with HIV-RNA levels below 50 copies/ml at week 48 (67%) is consistent with recent results involving antiretroviral-naive patients receiving two NRTI plus either boosted fosamprenavir [18] or boosted atazanavir [19]. In our trial, the ITT-E analysis penalized the LPV/r triple therapy arm compared with the LPV/r monotherapy arm. At week 48, 16 and 23% of patients discontinued the study for any reason on LPV/r monotherapy and LPV/r triple therapy, respectively. Such a percentage on LPV/r triple therapy is concordant with results found in a recent trial also involving naive patients receiving a similar regimen [18]. Therefore patients receiving LPV/r triple therapy were almost exclusively classified as non-responders as a result of discontinuation; whereas on LPV/r monotherapy, non-response resulted from a mixture of virological failures and discontinuations.
A higher proportion of intermittent viraemia (HIV RNA > 50 but < 400 copies/ml) was seen in the LPV/r monotherapy arm (Fig. 2), not only in this study with antiretroviral-naive patients, but also in maintenance studies in patients with previously undetectable viral loads [20,21]. Prolonged periods of low-level viraemia might favour the development of resistance mutations, as seen in the three patients on LPV/r monotherapy in our study, who developed resistance mutations after several weeks of almost continuous low-level viraemia (Fig. 3). The cause of these episodes of intermittent viraemia is not clearly understood at present. One possible explanation for the intermittent viraemia seen with LPV/r monotherapy is that in the absence of NRTI and their longer intracellular half-life, greater adherence may be required for constant suppression with LPV/r monotherapy. Of note is the fact that the median lopinavir trough concentration was similar between the two randomly assigned groups, although a limitation is that lopinavir trough measurements were recorded only at three timepoints. We observed, however, a correlation between the lopinavir trough concentration and self-reported adherence.
The long-term durability of the antiviral activity of LPV/r with two NRTI has been confirmed [11,12]. Studies with LPV/r in combination with two NRTI showed that the high inhibitory quotient of LPV/r at any time yields a plasma concentration of lopinavir that allows little replication of HIV [22-24]. Moreover, the high genetic barrier of LPV/r against resistance generally requires the accumulation of a high number of protease mutations to confer resistance [25]. Additional mutations that were not present at baseline were detected in the protease gene in three of 83 patients (3.6%) on LPV/r monotherapy, and in the reverse transcriptase gene in one patient on LPV/r triple therapy up to week 48. The absolute risk of resistance over 48 weeks with LPV/r monotherapy does not appear to be high compared with triple non-nucleoside reverse transcriptase inhibitor-based therapy among antiretroviral-naive patients; however, it appears to be higher than has been observed with triple therapy protease inhibitor-associated resistance regimens [18,24]. The mutations selected yielded neither significant phenotypic nor genotypic resistance to LPV/r at the time of failure. In the context of this closely monitored clinical trial, however, the resistance observed does raise caution regarding any widespread usage of LPV/r monotherapy in an antiretroviral-naive population. Whether these results might be extended to the whole protease inhibitor-associated resistance drug class remains to be determined [26].
Importantly, in the LPV/r monotherapy arm, three patients (of which only one experienced a protocol-defined suboptimal response) were intensified with zidovudine and lamivudine, and all three achieved prolonged full viral suppression. This is in keeping with previous LPV/r maintenance therapy studies, in which most patients experiencing virological failure on LPV/r achieved full viral suppression with the addition of their previous NRTI backbone [20,21]. Therefore, in this study, virological failure on LPV/r monotherapy did not jeopardize future therapeutic options over the observed duration of follow-up. Long-term follow-up of LPV/r monotherapy patients will be required to assess the risks relative to resistance over time.
The limitations of this study include its open-label design, which might bias an assessment of tolerability or toxicity, but would be less likely to bias virological, immunological, or resistance endpoints. Another limitation may be the pilot nature of the study, with the limited power and the unbalanced sample size designed to focus on the LPV/r monotherapy arm. Finally, another limitation is that the study employed conservative entry criteria for both HIV-RNA level (< 100 000 copies/ml) and CD4 cell count (> 100 cells/μl). Therefore, results from this study may not be generalizable to the HIV-infected patient population as a whole.
Interestingly, 90% of patients randomly assigned to the LPV/r monotherapy group with HIV RNA below 400 copies/ml at week 4 reached the protocol-defined virological response. Therefore, an early decrease in the plasma viral load below 400 copies/ml might enable clinicians to predict potential responders to LPV/r monotherapy. We conclude that first-line monotherapy with LPV/r soft gel capsules is virologically less effective than the current standard-of-care triple combination with two NRTI and LPV/r soft gel capsules. Given the requirement for chronic therapy with current antiretroviral treatments, however, and the long-term toxicities associated with all antiretroviral therapies, long-term strategies that limit exposure while providing adequate virological efficacy deserve further study. Future monotherapy studies utilizing the more convenient LPV/r tablet formulation should focus on select patient populations, such as those with extensive reverse transcriptase inhibitor resistance (for example in developing countries where first to second-line switch is often driven by clinical failure and results in extensive NRTI resistance) [27]. Taking into account the long-term rates of lipoatrophy, viral resistance, patient satisfaction and the cost of therapy are also critical to identify clinical scenarios in which LPV/r monotherapy might yet play a significant role in the treatment of HIV infection.
Introduction
HAART has dramatically improved the prognosis of HIV disease [1]. Currently, the gold standard treatment is a combination regimen that typically includes two nucleoside analogue reverse transcriptase inhibitors (NRTI) plus either one protease inhibitor or one non-nucleoside reverse transcriptase inhibitor. The absence of HIV eradication with those drugs, however, requires their prolonged use for a lifespan, making their long-term cost and cumulative toxicities a critical issue in the treatment of HIV infection. Mitochondrial toxicity and lipoatrophy are well documented adverse effects of NRTI [2-6]. Lipoatrophy in the buttocks, limbs, or face are common and distressing side effects for patients receiving anti-HIV therapy, and can lead to a reduction in patient adherence to therapy [7].
It has been hypothesized that monotherapy with a ritonavir-boosted HIV protease inhibitor may offer protection from the long-term toxicities of NRTI while effectively maintaining long-term virological suppression. Different strategies of ritonavir-boosted protease inhibitor monotherapy have been considered. One strategy discontinues the NRTI backbone in patients with full viral suppression on a classic triple combination with two NRTI and one ritonavir-boosted protease inhibitor, and recent results suggest that these strategies are effective [8-10]. Another approach is to use ritonavir-boosted protease inhibitor monotherapy initially as a first line regimen, thereby avoiding NRTI exposure entirely. Lopinavir/ritonavir (LPV/r) monotherapy is hypothetically well suited for evaluation in monotherapy trials on the basis of its virological potency and apparent high barrier to resistance development when employed as first-line therapy [11,12]. Although some pilot studies have suggested that first-line ritonavir-boosted protease inhibitor monotherapy seems effective in some specific populations [13], by the start of the study in October 2003, no randomized study had clearly challenged the notion that a first-line three-drug regimen is a prerequisite for successful anti-HIV therapy. Our objective was therefore to compare the safety and efficacy of a first-line monotherapy with LPV/r with a recommended classic triple regimen combining zidovudine plus lamivudine and LPV/r [14].
Results
Patient characteristics at baseline and patient disposition
Patients were randomly assigned into the study between October 2003 and February 2005. A total of 213 patients was screened for inclusion in the study (Fig. 1). The ITT-E population included 83 and 53 patients randomly assigned to LPV/r monotherapy and LPV/r triple therapy, respectively, who received their allocated treatment. Baseline characteristics were well balanced between the two arms (Table 1). More women, however, tended to be randomly assigned to LPV/r triple therapy (P = 0.10). Thirteen (16%) and twelve (23%) patients randomly assigned to LPV/r monotherapy and LPV/r triple therapy, respectively, discontinued the study before the 48-week visit (Fig. 1). The rate of discontinuation (P = 0.37) and time to discontinuation (log-rank test, P = 0.25) were not significantly different between the two arms, although a larger percentage of patients on LPV/r triple therapy tended to discontinue the study earlier than those on LPV/r monotherapy.
Virological and immunological response
In the monotherapy arm, seven patients (8.4%) had a confirmed viral rebound greater than 1.0 log10. Similarly, suboptimal virological response was reported in seven patients (13%) on LPV/r triple therapy: two patients failed to achieve a decline in HIV RNA of 1.0 log10 or greater by week 4, two patients failed to reach an HIV-RNA level of less than 400 copies/ml by week 24, and three patients had a confirmed viral rebound greater than 1.0 log10 after an HIV-RNA level of less than 400 copies/ml. There was no significant difference between the two arms (P = 0.40). By ITT-E analysis, 53 patients (64%) on LPV/r monotherapy and 40 patients (75%) on LPV/r triple therapy reached the study primary endpoint at week 48 (P = 0.19, Table 2). The point estimate of the difference between the two groups was 11.6% with a 95% confidence interval of -0.04 to 27%. In the on-treatment analysis, the proportion of responders was significantly lower on LPV/r monotherapy than on LPV/r triple therapy (80 versus 98%, respectively, P = 0.02). The difference between the two groups was 17.3% and the 95% confidence interval was 7-28%. Furthermore, fewer patients on LPV/r monotherapy had an HIV-RNA level of less than 50 copies/ml at week 48 compared with those on LPV/r triple therapy (84 versus 98%, respectively, P = 0.03). Treatment response according to study entry HIV-1-RNA levels, discontinuations, missing HIV-1-RNA data and treatment change at all study visits is shown in Fig. 2.
In LPV/r monotherapy, the on-treatment analysis indicates that a baseline HIV-1-RNA decrease in viral load at week 1, and HIV-1-RNA levels both at weeks 2 and 4 were predictive risk factors of the primary endpoint (P < 0.10, Table 3). Among these variables only early virological response at week 4 was an independent significant predictor of achieving the study primary endpoint. Thirty-seven of 41 patients (90.2%) with HIV-1 RNA less than 400 copies/ml at week 4 reached the primary endpoint.
There was no significant difference between the two groups with respect to the median change in the CD4 cell count from baseline to week 48 (P = 0.65). At week 48, the median change (interquartile range) was +151 (+105 to +239) and +159 (+112 to +250) cells/μl in the LPV/r monotherapy and LPV/r triple therapy arms, respectively.
Seventeen (20%) and 15 (28%) patients in the LPV/r monotherapy and LPV/r triple therapy arms, respectively, declared having missed at least two doses during the nine evaluations of the adherence from baseline to week 48. Overall, 34, 48 and 63% of patients who declared having missed none, one or more than two doses, respectively, had at least one low (< 3000 ng/ml) lopinavir trough concentration (test for trend, P = 0.003). Similarly, at least one undetectable value of lopinavir trough concentration was observed in 4, 6, and 22% of patients who declared having missed none, one or more than two doses, respectively (test for trend, P = 0.004). Plasma lopinavir trough concentrations were in the same range in the responder and in the non-responder groups in the monotherapy arm. No therapeutic change decisions were made in any of these patients based on plasma lopinavir concentrations.
Genotypic resistance
Of the 24 patients who qualified for resistance testing, the majority (21/24) received LPV/r monotherapy. Protease inhibitor-associated resistance mutations emerged in three of 21 patients on LPV/r monotherapy (L76V, M46I). Phenotypic changes in lopinavir sensitivity were either not observed or were extremely modest ranging from a 1.13 to 2.69-fold increase compared with reference (Fig. 3). Trough plasma lopinavir concentrations were always above 3500 ng/ml in these three patients (Fig. 3). An NRTI-associated resistance mutation was seen in one of three patients on LPV/r triple therapy who required genotypic testing (M184V at week 24 in patient no. 5102); no patient on triple therapy developed protease inhibitor resistance mutations.
Tolerance
Ten of 83 patients (12%) on LPV/r monotherapy and four of 53 patients (8%) on LPV/r triple therapy experienced a serious adverse event throughout the study. None of the serious reported adverse events was considered to be related to study treatment by the investigator on site. Adverse events of at least moderate severity occurred with similar frequency in the two treatments groups. Diarrhoea was the most common clinical adverse event [five (6%) and four (8%) patients on LPV/r monotherapy and LPV/r triple therapy, respectively], whereas aspartate aminotransferase and alanine aminotransferase elevation was the most common laboratory abnormality [10 (12%) and four (8%) patients on LPV/r monotherapy and LPV/r triple therapy, respectively].
Discussion
This is the first randomized trial comparing LPV/r monotherapy as a first-line regimen with a recommended potent triple regimen of zidovudine, lamivudine and LPV/r [15]. In these antiretroviral-naive patients, using ITT-E analysis, 64% of patients randomly assigned reached the protocol-defined virological response in the LPV/r monotherapy arm compared with 75% in the LPV/r triple therapy arm. This difference was not statistically significant. The on-treatment analysis at week 48, however, shows a larger discrepancy between the two randomized groups with 80 and 98% of patients reaching the primary endpoint in the LPV/r monotherapy arm versus the LPV/r triple therapy arm. LPV/r monotherapy resulted in similar CD4 cell count increases as obtained with the standard triple drug combination. Patients on LPV/r monotherapy reported significantly fewer symptoms than those on triple combination.
In the LPV/r triple therapy arm, by ITT-E analysis, the percentage of patients with HIV-RNA levels below 50 copies/ml at week 48 (67%) is consistent with recent results involving antiretroviral-naive patients receiving two NRTI plus either boosted fosamprenavir [18] or boosted atazanavir [19]. In our trial, the ITT-E analysis penalized the LPV/r triple therapy arm compared with the LPV/r monotherapy arm. At week 48, 16 and 23% of patients discontinued the study for any reason on LPV/r monotherapy and LPV/r triple therapy, respectively. Such a percentage on LPV/r triple therapy is concordant with results found in a recent trial also involving naive patients receiving a similar regimen [18]. Therefore patients receiving LPV/r triple therapy were almost exclusively classified as non-responders as a result of discontinuation; whereas on LPV/r monotherapy, non-response resulted from a mixture of virological failures and discontinuations.
A higher proportion of intermittent viraemia (HIV RNA > 50 but < 400 copies/ml) was seen in the LPV/r monotherapy arm (Fig. 2), not only in this study with antiretroviral-naive patients, but also in maintenance studies in patients with previously undetectable viral loads [20,21]. Prolonged periods of low-level viraemia might favour the development of resistance mutations, as seen in the three patients on LPV/r monotherapy in our study, who developed resistance mutations after several weeks of almost continuous low-level viraemia (Fig. 3). The cause of these episodes of intermittent viraemia is not clearly understood at present. One possible explanation for the intermittent viraemia seen with LPV/r monotherapy is that in the absence of NRTI and their longer intracellular half-life, greater adherence may be required for constant suppression with LPV/r monotherapy. Of note is the fact that the median lopinavir trough concentration was similar between the two randomly assigned groups, although a limitation is that lopinavir trough measurements were recorded only at three timepoints. We observed, however, a correlation between the lopinavir trough concentration and self-reported adherence.
The long-term durability of the antiviral activity of LPV/r with two NRTI has been confirmed [11,12]. Studies with LPV/r in combination with two NRTI showed that the high inhibitory quotient of LPV/r at any time yields a plasma concentration of lopinavir that allows little replication of HIV [22-24]. Moreover, the high genetic barrier of LPV/r against resistance generally requires the accumulation of a high number of protease mutations to confer resistance [25]. Additional mutations that were not present at baseline were detected in the protease gene in three of 83 patients (3.6%) on LPV/r monotherapy, and in the reverse transcriptase gene in one patient on LPV/r triple therapy up to week 48. The absolute risk of resistance over 48 weeks with LPV/r monotherapy does not appear to be high compared with triple non-nucleoside reverse transcriptase inhibitor-based therapy among antiretroviral-naive patients; however, it appears to be higher than has been observed with triple therapy protease inhibitor-associated resistance regimens [18,24]. The mutations selected yielded neither significant phenotypic nor genotypic resistance to LPV/r at the time of failure. In the context of this closely monitored clinical trial, however, the resistance observed does raise caution regarding any widespread usage of LPV/r monotherapy in an antiretroviral-naive population. Whether these results might be extended to the whole protease inhibitor-associated resistance drug class remains to be determined [26].
Importantly, in the LPV/r monotherapy arm, three patients (of which only one experienced a protocol-defined suboptimal response) were intensified with zidovudine and lamivudine, and all three achieved prolonged full viral suppression. This is in keeping with previous LPV/r maintenance therapy studies, in which most patients experiencing virological failure on LPV/r achieved full viral suppression with the addition of their previous NRTI backbone [20,21]. Therefore, in this study, virological failure on LPV/r monotherapy did not jeopardize future therapeutic options over the observed duration of follow-up. Long-term follow-up of LPV/r monotherapy patients will be required to assess the risks relative to resistance over time.
The limitations of this study include its open-label design, which might bias an assessment of tolerability or toxicity, but would be less likely to bias virological, immunological, or resistance endpoints. Another limitation may be the pilot nature of the study, with the limited power and the unbalanced sample size designed to focus on the LPV/r monotherapy arm. Finally, another limitation is that the study employed conservative entry criteria for both HIV-RNA level (< 100 000 copies/ml) and CD4 cell count (> 100 cells/μl). Therefore, results from this study may not be generalizable to the HIV-infected patient population as a whole.
Interestingly, 90% of patients randomly assigned to the LPV/r monotherapy group with HIV RNA below 400 copies/ml at week 4 reached the protocol-defined virological response. Therefore, an early decrease in the plasma viral load below 400 copies/ml might enable clinicians to predict potential responders to LPV/r monotherapy. We conclude that first-line monotherapy with LPV/r soft gel capsules is virologically less effective than the current standard-of-care triple combination with two NRTI and LPV/r soft gel capsules. Given the requirement for chronic therapy with current antiretroviral treatments, however, and the long-term toxicities associated with all antiretroviral therapies, long-term strategies that limit exposure while providing adequate virological efficacy deserve further study. Future monotherapy studies utilizing the more convenient LPV/r tablet formulation should focus on select patient populations, such as those with extensive reverse transcriptase inhibitor resistance (for example in developing countries where first to second-line switch is often driven by clinical failure and results in extensive NRTI resistance) [27]. Taking into account the long-term rates of lipoatrophy, viral resistance, patient satisfaction and the cost of therapy are also critical to identify clinical scenarios in which LPV/r monotherapy might yet play a significant role in the treatment of HIV infection.
|
|
| |
| |
|
|
|
