Comparative studies of two-times-daily versus three-times-daily indinavir in combination with zidovudine and lamivudine

Objectives: To compare the efficacy and safety of two-times-daily versus three-times-daily indinavir in combination with zidovudine and lamivudine.

Cited Here...: Two multicenter, open-label, randomized 24-week studies.

Cited Here...: Adults HIV-1 infection, HIV-1 RNA greater than 10 000 copies/ml, and no prior lamivudine or protease inhibitor therapy were eligible. In a pilot study (Study A), patients received indinavir at 800 mg every 8 h, 1000 mg every 12 h, or 1200 mg every 12 h. In a subsequent study (Study B), patients received indinavir at 800 mg every 8 h or 1200 mg every 12 h. All subjects received zidovudine (300 mg) and lamivudine (150 mg) every 12 h. An intent-to-treat analysis was used.

Cited Here...: In Study A, which enrolled 88 patients, neither HIV-1 RNA nor CD4 cell responses differed significantly between treatment groups at 24 weeks when corrected for multiple comparisons. Study B enrolled 433 patients, but was prematurely discontinued when interim analysis suggested greater efficacy of three-times-daily indinavir. Of the first 87 patients reaching week 24, HIV-1 RNA was less than 400 copies/ml in 91% receiving three-times-daily versus 64% receiving two-times-daily indinavir (P < 0.01).

Conclusion: Three-times-daily indinavir appears more efficacious than two-times-daily dosing when administered with zidovudine and lamivudine. Two-times-daily indinavir dosing should only be considered in situations characterized by favorable pharmacokinetic drug-drug interactions.

Introduction

Controlling HIV-1 replication with antiretroviral therapy reduces morbidity and mortality, and optimal ways to achieve durable inhibition of HIV-1 replication are being defined by comparative clinical trials ^[1-14]. For patients naïve to protease inhibitors and lamivudine, indinavir administered every 8 h plus twice-daily zidovudine and lamivudine controls HIV-1 replication ^[15-17] but requires three-times-daily indinavir dosing.

Earlier clinical trials suggested that no single indinavir plasma pharmacokinetic parameter predicted maximal efficacy. Daily indinavir area under the curve and trough values were comparable between patients achieving HIV-1 RNA levels less than versus greater than 500 copies/ml ^[18], suggesting that indinavir could be effectively administered twice daily at a full daily dose. Two randomized studies compared two-times-daily versus three-times-daily indinavir administered in combination with two-times-daily zidovudine and lamivudine. This article presents results from all patients in the pilot study, and from the first 287 patients completing at least 16 weeks of the subsequent study before its premature termination.

Materials and methods

Study design and patients

Study A was a pilot, multicenter, open-label, randomized trial which randomized patients in equal numbers to indinavir at 800 mg every 8 h, 1000 mg every 12 h, or 1200 mg every 12 h. The objectives were to rule out large between-group differences in the proportions of patients achieving serum HIV-1 RNA levels below 500 copies/ml and 50 copies/ml at 24 weeks, absolute changes in HIV-1 RNA and CD4 cell counts, and drug-related adverse experiences. Broader goals were to determine whether to initiate the larger study (Study B), and with which dose. Patients were enrolled at eight sites in the United States and Puerto Rico. Initially the protocol required prior nucleoside reverse transcriptase inhibitor experience, but was amended to include antiretroviral-naive patients. With 90 patients, Study A had an 80% power to detect a 35 percentage-point difference between groups regarding proportion with HIV-1 RNA below 500 copies/ml at 24 weeks.

Study B was a multicenter, open-label trial which randomized protease inhibitor and lamivudine-naive patients to indinavir at 800 mg every 8 h or 1200 mg every 12 h. The objective was to determine whether the antiviral activity, safety, and tolerability of indinavir at 1200 mg every 12 h was equivalent to 800 mg every 8 h when co-administered with zidovudine and lamivudine for 24 weeks. Patients were enrolled at 49 sites in South America, North America, Europe, Australia, and New Zealand. The original protocol was subsequently amended to include an interim analysis when 24-week data became available on all patients enrolled by 30 April 1998. If the true proportion for each group was 70%, approximately 310 patients per group would establish equivalence with 90% power, assuming a Pocock α-adjustment for the interim look ^[19]. The sample size for study B was determined on the premise that equivalence would be established if the upper endpoint of the α-adjusted 95% confidence interval for the between group difference in the proportion of patients with serum HIV-1 RNA less than 400 copies/ml at 24 weeks was less than 12 percentage points.

All patients received concomitant zidovudine (300 mg every 12 h) and lamivudine (150 mg every 12 h). Eligible subjects were at least 18 years of age, with HIV-1 RNA greater than 10 000 copies/ml, and CD4 cell counts either between 150 and 500 × 10⁶ cells/l (Study A) or greater than 100 × 10⁶ cells/l (Study B).

Evaluations performed at baseline and at least every 4 weeks included HIV-1 RNA assays [Roche (Nutley, New Jersey, USA) Amplicor1 method, quantification limit 500 copies/ml during Study A and 400 copies/ml in Study B, and Roche Ultrasensitive Assay, quantification limit 50 copies/ml], CD4 cell counts, routine hematology, chemistry, coagulation studies, and urinalyses. Indinavir (Crixivan) 200 mg capsules, zidovudine (Retrovir) 100 mg capsules, and lamivudine (Epivir) 150 mg tablets were supplied by Merck and Co., Inc. (West Point, Pennsylvania, USA). Studies were approved by the institutional review boards, and patients gave informed consent.

Statistics

Efficacy data were analyzed using intent-to-treat model-based methods. This accounts for all randomized patients, and reduces bias introduced by premature discontinuations. In Study B, only patients who could have been on study for at least 16 weeks were included in the analyses. For the proportion of patients achieving HIV-1 RNA levels below limits of quantification, the generalized estimating equations (GEE) methodology for longitudinal data analysis ^[20] was used after explicitly imputing HIV-1 RNA failures from the time of discontinuation onward for all treatment-related dropouts (e.g. adverse experience) regardless of HIV-1 RNA level. Similarly, for HIV-1 RNA and CD4 cell change from baseline, the restricted maximum likelihood (REML) methodology for longitudinal data analysis ^[21] was used after carrying forward the last observation for all treatment related dropouts. For the latter analyses, baseline values were used as covariates (for study A), and values of 250 (or 200 for study B) copies/ml and 500 (or 400 for study B) copies/ml were assigned for HIV-1 RNA assays with no amplifiable signal and detectable signal below 500 (or 400 for Study B) copies/ml (limit of quantification), respectively. In Study A, comparisons of the three-times-daily group to each two-times-daily group were performed after adjusting for multiple comparisons ^[22].

The incidence of adverse experiences, and of premature discontinuations were compared between groups using pairwise two-tailed Fisher's exact tests, without adjustments for multiple comparisons.

Results

Patient treatment and follow-up

Study A enrolled 88 patients. Treatment groups were well matched for CD4 cell counts and HIV-1 RNA levels, although prior antiretroviral experience was slightly less in the three-times-daily (79%) versus the two-times-daily (93%, pooled) groups (P > 0.05, Table 1). Overall, 31 (35.2%) of 88 subjects discontinued the study before 24 weeks, with clinical adverse experience cited as the reason as the reason in 16 (51.6%) of these 31 subjects. At 24 weeks, 17 (58.6%) of patients in the three-times-daily group, 22 (75.9%) in the 1000 mg two-times-daily group, and 18 (60.0%) in the 1200 mg two-times-daily group remained on study (P > 0.05).

Table 1

Study B enrolled 443 of the planned 620 patients by 30 April 1998. At the time of interim/final analysis, 287, 171, and 87 patients could potentially have been on study for at least 16, 20, and 24 weeks, respectively. These 287 subjects were well matched for CD4 cell counts, HIV-1 RNA levels, and prior antiretroviral therapy (Table 1). Overall, 41 (14.3%) of these 287 patients discontinued before 24 weeks, with clinical adverse experience cited as the reason in 14 (34.1%) of the 41. At interim/final analysis, 124 (87.3%) in the three-times-daily group and 122 (84.1%) in the two-times-daily group remained on study (P > 0.05). The 156 patients who could not have reached week 16 at the time of interim analysis are excluded from subsequent analyses.

HIV-1 RNA responses

In Study A, HIV-1 RNA decreased in all three groups. The proportions of patients with HIV-1 RNA less than 500 copies/ml at 24 weeks were 36.2, 62.3, and 53.7% in the three-times-daily, 1000 mg two-times-daily, and 1200 mg two-times-daily groups, respectively. The percentages of patients achieving HIV-1 RNA levels below 50 copies/ml at 24 weeks were 28.9, 49.4, and 46.2%, respectively. At 24 weeks, mean decreases from baseline in HIV-1 RNA for the three-times-daily, 1000 mg two-times-daily, and 1200 mg two-times-daily groups were 1.43, 1.61, and 1.68 log₁₀ copies/ml, respectively. The data favoring two-times-daily regimens was not significant when adjusted for multiple comparisons (P > 0.05).

In Study B, the proportion of patients achieving HIV-1 RNA levels less than 400 copies/ml in the three-times-daily group at 16, 20, and 24 weeks were 77.5, 89.9, and 90.8%, respectively. In the two-times-daily group these proportions were 71.6, 67.0, and 63.7%, respectively (Fig. 1a). The more sensitive HIV-1 RNA assay similarly favored three-times-daily indinavir (Fig. 1b). Mean HIV-1 RNA decreases at 24 weeks were 2.29 log₁₀ copies/ml in the three-times-daily group and 1.99 log₁₀ copies/ml in the two-times-daily group.

Fig. 1

CD4 lymphocytes responses

There were significant CD4 cell increases in all treatment groups. In Study A, mean CD4 cell count increases at 24 weeks in the 800 mg three-times-daily, 1000 mg two-times-daily, and 1200 mg two-times-daily groups were 116, 70, and 48 × 10⁶ cells/l, respectively (P > 0.05). In Study B, mean CD4 cell count increases at 24 weeks were 114 × 10⁶ cells/l in the three-times-daily group and 116 × 10⁶ cells/l in the two-times-daily group (adjusted P > 0.05).

Adverse events

In Study A there were no significant differences among the three groups regarding drug-related adverse events. The most common adverse experiences included nausea (61%), vomiting (33%), and asthenia/fatigue (26%). Nausea was generally mild, and only 28% of patients with nausea withdrew from the study. Six serious adverse experiences judged to be at least possibly drug-related included two (pyelonephritis and nephrolithiasis) in the three-times-daily group, one (diarrhea) in the 1000 mg two-times-daily group, and three (septicemia, nephrolithiasis, and diverticulitis) in the 1200 mg two-times-daily group. There were two cases of nephrolithiasis in the three-times daily group, one in the 1000 mg two-times-daily group, and four in the 1200 mg two-times-daily group (P > 0.05).

In Study B, numbers of premature discontinuations were comparable between groups: 18 (13%) of 142 in the three-times-daily group, and 23 (16%) of 145 in the two-times-daily group. However, of the 287 patients, 11 in the two-times-daily group versus three in the three-times-daily group discontinued for clinical adverse events, particularly nausea/vomiting (P > 0.05). In the two-times-daily group eight discontinued with nausea/vomiting, one with nephrolithiasis, one with hematuria/crystalluria, and one with paresthesias/headache. In the three-times-daily group, two discontinued with nausea/vomiting and one with flushing.

Discussion

These studies compared indinavir at 800 mg three-times-daily with 1200 mg (or 1000 mg) two-times-daily, in a three-drug regimen with zidovudine and lamivudine. Although the pilot study suggested that a higher proportion of patients receiving two-times-daily indinavir achieved HIV-1 RNA levels below both 500 and 50 copies/ml, the converse occurred in the larger study. This discrepancy may reflect differences in adherence between studies and the large proportion of discontinuations due to nausea/vomiting in the two-times-daily group in the larger study. Although the rationale for selecting 1200 mg rather than 1000 mg two-times-daily for study B was to enhance plasma drug levels, the high dose may have been less well tolerated.

The performance of the two-times-daily regimen was similar between the two studies. The major difference between studies related to the three-times-daily regimen. The three-times-daily regimen performed extremely well in the larger study, with over 90% of patients achieving HIV-1 RNA levels below 400 copies/ml at 24 weeks. Although there has some been variation in the degree of antiretroviral suppression among previous studies ^{[15-17,23,24]}, results similar to the present study B have been documented in studies of motivated populations ^[16,17,24]. It is unclear why the proportion of patients in Study A achieving HIV-1 RNA levels less than 500 copies/ml in the three-times-daily indinavir arm was less than in previous trials which enrolled zidovudine-experienced patients ^[15,16]. In ACTG 320, which enrolled patients with less than 200 × 10⁶ cells/l, 60% of such patients achieved plasma HIV-1 RNA levels less than 500 copies/ml at 24 weeks ^[15]. The higher baseline CD4 cell counts in study B as compared to ACTG 320 may explain the somewhat better virologic response to three-times-daily indinavir in the present study B. In Merck Protocols 035 and 060, which enrolled patients with less than 400 × 10⁶ CD4 cells/l and greater than 500 × 10⁶ CD4 cells/l, respectively, approximately 90% of subjects achieved this response at 24 weeks ^[16,24]. By contrast, only 36.2% of patients in study A achieved this virologic response.

It is not clear why three-times-daily indinavir was more efficacious than twice-daily dosing. Although three-times-daily dosing may have yielded a more favorable pharmacokinetic profile, we cannot exclude decreased adherence with the twice-daily regimen, perhaps due to intolerance. The safety profile of the two-times-daily indinavir regimen was similar to that for three-times-daily dosing, and was not associated with significant increases in nephrolithiasis. Future studies of less than three-times-daily indinavir should focus on favorable pharmacokinetic interactions, such as those with indinavir and ritonavir co-administration ^[25], as well as improving gastrointestinal tolerance to optimize adherence.

Acknowledgements

The authors are grateful to the persons with HIV infection who participated in this study, and to the following study site investigators and personnel: Willy Rozenbaum, Michael Dube, David Barker, Jefferey Fessel, Roy Steibigel, Charles Carpenter, William J. Holloway, Princy Kumar, Marla Gold, William Powderly, Kate Squires, Deborah McMahon, Rod Ellis-Peglar, Chris Tsoukas, Max Kronawetter, Norbert Vetter, Anne Phillips, Kevin Gough, Frank D. Goebel, Andreas Plettenberg, Bernard Vandercam, Neil Bodsworth, Julian Gold, Ferdinando Aiuti, Margaret Johnson, Nathan Clumeck, Anita Rachlis, John Gill, John MacLeod, Julio Montaner, Emilio Bouza, Pompeyo Viciana, Enrique Ortega, Javier Santamaria, Jose Mallolas, Mauro Moroni, C. Jeffrey Goodgame, G. Steven Kooshian, Claudia Ochoa, Janet Nicotera, Beverly Weaver, Barbara Sepcie, Andreia F. Colombo Barbosa, Greyce Lousanna, Rosa Lluveras, Charles Raines, Sydia Cruz, Deborah Johnson, Andrea Malebran, Irene Teran, and Dana Cummings.