If Taken 1 Hour before Indinavir (IDV), Didanosine Does Not Affect IDV Exposure, despite Persistent Buffering Effects

Abstract

Concurrent administration of indinavir and didanosine significantly reduces the level of exposure to indinavir, but it is unclear how soon after didanosine administration indinavir may be given safely. We compared indinavir pharmacokinetics and gastric pH in 12 human immunodeficiency virus-positive patients by use of 800 mg of indinavir alone versus 800 mg of indinavir administered 1 h after didanosine administration. Median gastric pH was significantly higher when indinavir was taken after didanosine administration; however, no significant difference in the maximum concentration in plasma or the area under the concentration-time curve from time zero to 8 h was observed. Indinavir may be taken with a light meal 1 h following the administration of 400 mg of didanosine.

The human immunodeficiency virus (HIV) protease inhibitor indinavir is often used in combination with didanosine, a nucleoside analogue reverse transcriptase inhibitor. Acidic gastric pH is thought to be necessary for optimum absorption of indinavir (Crixivan package insert; Merck & Co., Inc., West Point, Pa.), but didanosine is rapidly degraded at an acidic pH. Therefore, didanosine tablets have a gastric acid buffering component (calcium carbonate and magnesium hydroxide) to prevent degradation (Videx package insert; Bristol-Myers Squibb, Princeton, N.J.). Two 100-mg doses of didanosine in tablet form are reported to increase the gastric pH to 8 to 9 in vivo (3). Concurrent administration of these agents reduces the level of indinavir exposure by 80% (V. Mummaneni, S. Kaul, and C. A. Knupp, Abstr. 26th Annu. ACCP Meet., J. Clin. Pharmacol. 37:858–878, abstr. 34), which is most likely due to buffer-induced changes in gastric pH. The manufacturer of indinavir recommends that these drugs be administered at least 1 h apart. However, the in vivo duration of gastric buffering from didanosine formulations is unknown, and it remains unclear how soon after didanosine administration indinavir may be given safely. The present study was undertaken to determine the effect of didanosine on gastric pH and indinavir pharmacokinetics when indinavir is taken 1 h following didanosine administration. HIV type 1 (HIV-1)-infected individuals between the ages of 18 and 65 years who were receiving indinavir as part of combination, antiretroviral therapy were recruited for the study. All subjects were required to have a fasting, minimum gastric pH of <3 on at least two occasions within 4 weeks of study entry. Subjects were not eligible for participation in the study if they had any acute medical problems or grade 3 or 4 laboratory abnormalities by AIDS Clinical Trials Group criteria (abnormal hemoglobin levels, absolute neutrophil counts, platelet counts, or serum creatinine, aspartate aminotransferase, alanine aminotransferase, alanine phosphatase, or bilirubin levels). Excluded medications included investigational agents, hepatic enzyme-inducing or -inhibiting agents, other HIV protease inhibitors, nonnucleoside reverse transcriptase inhibitors, and any agents affecting gastric pH or gastric emptying. The protocol was approved by the Institutional Review Board at the State University of New York at Buffalo, and informed consent was obtained from all subjects prior to participation in the study. In random order, subjects received 800 mg of indinavir in a fasting state and 800 mg of indinavir with a light breakfast 1 h after administration of a 400-mg dose of didanosine. Gastric pH was measured with a radiotelemetric pH-monitoring device (Heidelberg [Norcross, Ga.] pH monitoring system), a consumable, nondigestible capsule that transmits a continuous radio signal to an antenna located in a belt worn by the subject. The output of the device is a graph of pH versus time.

All subjects arrived at the clinic on the morning after an overnight fast of at least 8 h. Following a brief assessment of safety and adherence, an intravenous catheter was placed into an antecubital fossa. For the control treatment with indinavir alone, all subjects ingested a radiotelemetry capsule just prior to ingestion of two 400-mg capsules of indinavir (total volume of water consumed, 6 oz). For the didanosine study treatment, a 400-mg dose of didanosine (two 150-mg tablets plus one 100-mg tablet dissolved in 6 oz of water) was administered 1 h prior to indinavir administration. Following a light breakfast (dry toast with jelly, apple juice, coffee with skim milk), a radiotelemetry capsule and two 400-mg capsules of indinavir (total volume of water consumed, 6 oz) were ingested. Each study treatment was separated by a minimum of 4 days but by no more than 14 days. Following indinavir administration, whole-blood samples (7 ml) were withdrawn into heparin-containing vacuum tubes at the following times after indinavir dosing: 0, 0.5, 1, 1.5, 2, 3, 4, and 5 h. Plasma was harvested by centrifugation of the specimens at 2,000 × g for 10 min. The upper plasma layer was then transferred to storage vials, which were frozen at −20°C and within 48 h were transferred to −70°C, where they remained until they were assayed.

The concentrations of indinavir were measured by a validated, isocratic, reverse-phase high-performance liquid chromatographic assay procedure. Ratios of the peak height of indinavir to that of an internal standard were calculated and used to formulate a calibration curve of concentration (x) versus peak height ratio (y) by weighted (1/concentration squared), least-squares, linear regression. Correlation coefficients were ≥0.995. The calibration curve of this assay ranged from 50 to 10,000 ng/ml. The intra- and interassay variations for validations were <15% at 160, 800, and 4,000 ng/ml. The limit of quantification was 100 ng/ml (0.162 μM). Pharmacokinetic analysis was accomplished with WIN-NONLIN Professional, version 2.1, software (SCI Software, Lexington, Ky.) by noncompartmental methods. Peak concentrations in plasma (C_max) and the time to C_max (T_max) were determined by inspection of concentration-time curves for individual subjects. The elimination rate constant (k_el) was determined from the slope of the terminal phase of the concentration-versus-time profile by least-squares regression with a monoexponential decay function. The area under the curve (AUC) from time zero to 5 h (AUC_0–5) was estimated by the linear trapezoidal rule and was then extrapolated to 8 h (AUC_0–8) by using k_el. Indinavir pharmacokinetic parameters (C_max, k_el, AUC_0–8) were compared across study treatments by using analysis of variance for repeated measures. Each of these pharmacokinetic parameters was log transformed prior to analysis and was treated as a continuous variable. T_max values were compared by the Wilcoxon signed rank test. In addition, the 90% confidence intervals of the mean ratio (C_max and AUC_0–8) for indinavir alone versus indinavir after didanosine were calculated and subjected to U.S. Food and Drug Administration (FDA) bioequivalency testing criteria. The intrasubject median gastric pH values in the first 30 min after indinavir administration were compared by Friedman's rank test. A P value of <0.05 was used as a measure of statistical significance.

The study was conducted from September to December of 1998. Fourteen patients were enrolled, with 12 patients completing the study: 8 men, 4 women, 6 Caucasians, 5 African Americans, and 1 Hispanic. Two subjects had taken indinavir previously but were on a “drug holiday” at the time of the study. These patients took two 800-mg doses of indinavir 16 and 8 h prior to each pharmacokinetic evaluation to achieve steady state with regard to indinavir pharmacokinetics. No medical events thought to be due to study medications occurred during or after the study period. The means ± standard deviations for pharmacokinetic and gastric pH parameters are provided in Table 1. When administered alone, indinavir was rapidly absorbed following oral dosing, with a median C_max of 14.8 ± 6.8 μM, which was reached at 1.3 ± 0.9 h. The mean C_max of indinavir after didanosine treatment was 12.6 ± 2.7 μM, which was similar to that of indinavir given alone. However, the mean T_max of indinavir following didanosine treatment appeared to be marginally prolonged compared to that of indinavir given alone. The AUC_0–8 for indinavir alone was no different from that for indinavir after didanosine treatment, although the variability appeared to be greater during the study with treatment with indinavir alone (Fig. 1). The 90% confidence intervals of the mean ratios for AUC_0–8 and C_max were 0.83 to 1.15 and 0.70 to 1.19, respectively, meeting FDA bioequivalency standards for both. Indinavir was rapidly eliminated, with a median apparent elimination half-life of 1.1 h after both study treatments. The median gastric pH during the first 30 min after indinavir administration was significantly higher when indinavir was taken after didanosine treatment (2.9 ± 1.9 versus 1.7 ± 1.5; P = 0.004). The gastric pH was higher in the indinavir-didanosine arm for approximately 90 min after didanosine administration and a light meal.

TABLE 1.

Indinavir pharmacokinetic parameters and gastric pH values according to study treatment ^a

Treatment	Cmax (μM)	Tmax (h)	kel (h−1)	ACU0–5 (μmol · h/liter)	AUC0–8 (μmol · h/liter)	Gastric pHb
Indinavir without didanosine	14.8 ± 6.8	1.3 ± 0.9	0.621 ± 0.21	31.3 ± 15.5	35.0 ± 18.1	1.7 ± 1.5
Didanosine and then indinavir	12.6 ± 2.7	1.5 ± 0.4	0.667 ± 0.16	29.9 ± 7.2	33.1 ± 8.7	2.9 ± 1.9
P value	0.556c	0.304d	0.174c	0.900c	0.988c	0.004e

^aValues are means ± standard deviations for 12 patients. 

^bMedian gastric pH value during the first 30 min after indinavir administration. 

^cRepeated measures by analysis of variance. 

^dWilcoxon signed rank test. 

^eFriedman's signed rank test. 

FIG. 1.

AUC_0–8 for indinavir when indinavir (IDV) was taken without didanosine (ddI) and when indinavir was taken 1 h after administration of didanosine.

Concurrent use of both indinavir and didanosine represents a challenge to adherence with combination antiretroviral therapy due to the frequent dosing of indinavir, the fasting requirements for both drugs, and the need to separate the administration of these drugs from one another. Due to the acid-labile nature of didanosine, all present formulations contain or are mixed with buffers that prevent intragastric degradation. Each didanosine tablet is formulated to provide 50% of the buffer considered necessary to neutralize the gastric acid. Although originally licensed with a twice-daily dosing interval, didanosine has recently been approved by the FDA for once-daily administration and made available in a 200-mg tablet (Videx package insert; Bristol-Myers Squibb). Prior to the availability of this new strength, physicians often prescribed once-daily doses of didanosine using various combinations of the available tablets, such as two 150-mg tablets plus one 100-mg tablet or four 100-mg tablets. While the present study used a three-tablet, 400-mg dose of didanosine, which reflected the clinical practice standards at the time of the study, this dose can now be achieved with two 200-mg tablets and, thus, less buffer than that used in the present study. A light meal was used with the didanosine study treatment due to the observation that didanosine buffering persisted for more than 1 h after administration of a two-tablet dose in some of the patients in a previous study (3). Since the same light meal does not affect indinavir exposure (Crixivan package insert), we believed that such a meal might act as a stimulus for gastric acid secretion and potentially permit the avoidance of indinavir malabsorption. A reduced level of indinavir exposure has been associated with a suboptimal virologic response in HIV-infected patients (1). Gastric pH was significantly elevated at the time of indinavir dosing with a light breakfast in the present study, despite a 1-h interval since didanosine treatment. Although the pH returned to the baseline about 90 min after didanosine administration, it is unclear whether it was due to the study meal or waning of the buffering effects due to the buffers in the didanosine tablets.

Although the manufacturer of indinavir suggests that a normal gastric pH environment may be necessary for optimal indinavir absorption, the relatively minor increase in gastric pH that was present after didanosine administration did not appear to affect indinavir exposure. Despite the 5- to 6-h duration of sampling in the present study, the AUC_0–5/AUC_0–8 ratios were 91% ± 8.4% and 91% ± 4.6% for the control and didanosine study treatments, respectively. The pharmacokinetic parameters for 800 mg of indinavir reported in this study are similar to those noted previously (2, 4; Crixivan package insert [Merck & Co., Inc.]). Furthermore, statistical comparisons between study treatments for both AUC_0–5 and AUC_0–8 yielded the same results. Therefore, indinavir may be taken safely with a light breakfast 1 h following didanosine treatment.