ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Feb. 2009, p. 609–614
`0066-4804/09/$08.00⫹0 doi:10.1128/AAC.00769-08
`Copyright © 2009, American Society for Microbiology. All Rights Reserved.
`
`Vol. 53, No. 2
`
`Drug-Drug Interaction Study of Ketoconazole and
`Ritonavir-Boosted Saquinavir䌤
`Benoite Kaeser,1 Hagen Zandt,1 Fabrice Bour,3 Elke Zwanziger,1
`Christophe Schmitt,1 and Xiaoping Zhang2*
`Hoffmann-La Roche, Ltd., Basel, Switzerland1; Hoffman-La Roche, Inc., Nutley, New Jersey2; and
`Institut de Pharmacologie Clinique, Roche, Strasbourg, France3
`
`Received 12 June 2008/Returned for modification 28 September 2008/Accepted 8 November 2008
`
`Saquinavir, a potent human immunodeficiency virus protease inhibitor, is extensively metabolized by
`CYP3A4. Saquinavir is coadministered with ritonavir, a strong CYP3A4 inhibitor, to boost its exposure.
`Ketoconazole is a potent CYP3A inhibitor. The objectives of this study were to investigate the effect of
`ketoconazole on the pharmacokinetics of saquinavir/ritonavir and vice versa using the approved dosage
`regimens of saquinavir/ritonavir at 1,000/100 mg twice daily and ketoconazole at 200 mg once daily. This was
`an open-label, randomized two-arm, one-sequence, two-period crossover study in healthy subjects. In study
`arm 1, 20 subjects received saquinavir/ritonavir treatment alone for 14 days, followed in combination with
`ketoconazole treatment for 14 days. In arm 2, 12 subjects received ketoconazole treatment for 6 days, followed
`in combination with saquinavir/ritonavir treatment for 14 days. The pharmacokinetics were assessed on the
`last day of each treatment (days 14 and 28 in arm 1 and days 6 and 20 in arm 2). The exposures Cmax and the
`area under the concentration-time curve from 0 to 12 h (AUC0-12) of saquinavir and ritonavir with or without
`ketoconazole were not substantially altered after 2 weeks of concomitant dosing with ketoconazole. The Cmax
`and AUC0-12 of ketoconazole, dosed at 200 mg once daily, were increased by 45% (90% confidence interval ⴝ
`32 to 59%) and 168% (90% confidence interval ⴝ 146 to 193%), respectively, after 2 weeks of concomitant dosing
`with ritonavir-boosted saquinavir (1,000 mg of saquinavir/100 mg of ritonavir given twice daily). The greater
`exposure to ketoconazole when given in combination with saquinavir/ritonavir was not associated with unac-
`ceptable safety or tolerability. No dose adjustment for saquinavir/ritonavir (1,000/100 mg twice daily) is
`required when coadministered with 200 mg of ketoconazole once daily, and high doses of ketoconazole (>200
`mg/day) are not recommended.
`
`Saquinavir, a potent human immunodeficiency virus (HIV)
`protease inhibitor, is extensively metabolized by cytochrome
`P450 (CYP3A4; see the product information for Invirase cap-
`sules and tablets [Hoffman-La Roche, Inc.]). For the antiret-
`roviral combination treatment of HIV-infected patients,
`saquinavir is coadministered with ritonavir, another HIV pro-
`tease inhibitor and strong CYP3A4 inhibitor (see the product
`information for the Norvir 100-mg capsule [Abbott Laborato-
`ries]), to boost the exposure of saquinavir. The new dosage
`form of saquinavir (Invirase), a 500-mg film-coated tablet, was
`approved in 2004 in the United States and 2005 in Europe and
`allows a lower pill burden using the dosing regimen of saquina-
`vir/ritonavir at 1,000/100 mg twice daily, which was approved in
`2003. Saquinavir exhibits a pronounced food effect, and pa-
`tients are advised to take saquinavir with ritonavir always with
`a full meal (according to the product information for Invirase
`capsules and tablets [Hoffman-La Roche, Inc.]). The imidazole
`antifungal compound ketoconazole is effective in the treatment
`of oropharyngeal candidiasis, the most common infection
`among persons infected with HIV (1). Ketoconazole is also
`known as a prototypic and strong CYP3A4 inhibitor (2) and is
`a recommended drug of choice for investigating the interaction
`
`* Corresponding author. Mailing address: Department of Clinical
`Pharmacology, Hoffmann-La Roche, Inc., 340 Kingsland St., Nutley, NJ
`07110. Phone: (973) 235-6584. Fax: (973) 235-5635. E-mail: xiaoping
`.zhang@roche.com.
`䌤 Published ahead of print on 17 November 2008.
`
`with CYP3A4 substrates such as saquinavir (5). It has already
`been shown that coadministration of ketoconazole at 200 or
`400 mg once daily with saquinavir/ritonavir at 400/400 mg twice
`daily resulted in increased area under the drug plasma con-
`centration-time curve (AUC) and peak plasma concentrations
`(Cmax) for saquinavir and ritonavir, whereas the effect of
`saquinavir/ritonavir on the pharmacokinetics of ketoconazole
`was not evaluated (4). The present study was performed using
`the approved dosing regimens of saquinavir/ritonavir at 1,000/
`100 mg twice daily and ketoconazole at 200 mg once daily and
`documents the drug-drug interaction after 2 weeks of concom-
`itant dosing in both directions, i.e., the effect of ketoconazole
`on the pharmacokinetics of saquinavir/ritonavir and that of
`saquinavir/ritonavir on ketoconazole. The objective of the
`study was to provide appropriate dosing guidelines to clinicians
`who treat HIV patients with saquinavir/ritonavir and ketoco-
`nazole.
`
`MATERIALS AND METHODS
`
`Study design and population. This was an open-label, randomized two-arm,
`one-sequence, two-period crossover study conducted in healthy male and female
`subjects. Subjects had to give written informed consent, and the study protocol
`was approved by the IRB Institutional Review Board (Comite´ Consultatif de
`Protections des Personnes dans la Recherche Biome´dicale, Strasbourg, France).
`The study was conducted in full compliance with the principles of the Declara-
`tion of Helsinki III and performed according to the guidelines of Good Clinical
`Practice.
`In order to enable stable pharmacokinetic conditions for saquinavir/ritonavir,
`the combination of these two drugs was administered for 2 weeks, ketoconazole
`
`609
`
`1
`
`TEVA1037
`
`

`

`610
`
`KAESER ET AL.
`
`ANTIMICROB. AGENTS CHEMOTHER.
`
`alone was dosed for 6 days, and the combination of ritonavir/saquinavir with
`ketoconazole was administered in both study arms for another 2 weeks. In study
`arm 1, saquinavir/ritonavir was administered first (period 1), followed by the
`addition of ketoconazole (period 2), and in study arm 2, ketoconazole was dosed
`first (period 1), followed by the addition of saquinavir/ritonavir (period 2). The
`planned sample size for study arm 1 was n ⫽ 20, assuming a within-subject
`coefficient of variation (CV) of up to 30% for the saquinavir AUC and Cmax,
`which was based on the observed within-subject CV of 27 and 24% for these
`parameters, respectively, from a previous study (BP17359 [Roche data on file]).
`For study arm 2, the sample size was set to n ⫽ 12, assuming a within-subject CV
`of up to 16% for the ketoconazole AUC and Cmax based on the respective values
`of 16 and 14% observed in a previous study (WK14435 [Roche data on file]).
`These sample sizes would ensure that, with a probability of at least 80%, the
`two-sided 90% confidence interval (CI) for the geometric population mean of
`the individual parameter ratios (period 2/period 1) would be within 75 to 133%
`of the geometric population mean for saquinavir and within 80 to 125% of the
`geometric population mean for ketoconazole. Subjects were randomized to study
`arms 1 and 2 with a block size of eight (five to arm 1, three to arm 2).
`Subjects underwent screening evaluations to determine eligibility within 28
`days prior to study enrollment. Screening procedures included, among others,
`tests for HIV, tests for hepatitis B and C, and tests for pregnancy in women.
`Healthy male and female subjects, aged 18 to 65 years (inclusive) with body mass
`indexes between 18 to 30 kg/m2 and being nonsmokers, were enrolled into the
`study. Intake of grapefruit or grapefruit juice was not allowed from 2 weeks prior
`to the first dose and during the study. In addition, the consumption of alcohol
`was not permitted during the study. Subjects were instructed to take the study
`drugs always with a meal. No concomitant medications were permitted during
`the study except to treat adverse events. Subjects who were on concomitant
`treatment with drugs known as CYP3A4 substrates, CYP3A4 inhibitors, or
`CYP3A4 inducers were excluded from the study. Women in the study had to be
`of non-child-bearing potential or under efficient nonhormonal contraception
`throughout the study and until at least 1 month thereafter.
`In study arm 1, saquinavir/ritonavir at 1,000/100 mg was dosed twice daily for
`28 days (excluding the evening dose on day 14), with ketoconazole at 200 mg
`once daily added from days 15 to 28 (period 1, days 1 to 14; period 2, days 15 to
`28). In study arm 2, ketoconazole at 200 mg was dosed once daily for 20 days with
`saquinavir/ritonavir at 1,000/100 mg twice daily added from days 7 to 20 (period
`1, day 1 to 6; period 2, day 7 to 20). In study arm 1 the pharmacokinetics were
`assessed for saquinavir/ritonavir over 12 h on days 14 and 28, and in arm 2 the
`pharmacokinetics were assessed for ketoconazole over 24 h on days 6 and 20.
`Study drugs were administered 30 min after the start of a standard high-fat,
`high-calorie breakfast (63 g of fat, 975 kcal) on days 14 and 28 for arm 1 or days
`6 and 20 for arm 2. Plasma samples for pharmacokinetic assessments were
`collected predose and at 0.5, 1, 2, 3, 4, 5, 6, 8, 10, 12, 14, and 24 h postdose (13
`samples). In study arm 1, subjects received on days 14 and 28 saquinavir and
`ritonavir doses only in the morning. In study arm 1, plasma samples were
`analyzed for saquinavir and ritonavir, and in study arm 2 plasma samples were
`analyzed for ketoconazole. In both study arms additional predose concentration
`measurements were made on the last 2 to 4 days of periods 1 and 2 to document
`stable pharmacokinetic conditions. In study arms 1 and 2, during both pharma-
`cokinetic assessment periods, subjects were confined to the study center. Safety
`parameters included medical history, physical examinations, standard safety lab-
`oratory assessments (hematology, blood chemistry, and urinalysis), vital signs,
`and serial electrocardiograms recorded at prespecified time points throughout
`the trial, as well as urine analyses for drugs of abuse, alcohol breath tests, and
`urine pregnancy tests (in females only). A medical follow-up examination was
`conducted 15 to 21 days after the last dose of study drugs. All subjects were
`observed for adverse events during the entire study period.
`Bioanalysis. For the determination of saquinavir and ritonavir pharmacoki-
`netics, blood samples of 5.5 ml were collected by peripheral venous catheter or
`venipuncture into tubes containing lithium heparin as the anticoagulant, and for
`the pharmacokinetics of ketoconazole, blood samples of 2.6 ml were collected
`into EDTA-containing tubes. Plasma was separated by centrifugation for 10 min
`at 1,500 ⫻ g and 4°C. The total plasma concentrations of saquinavir and ritonavir
`were analyzed by PRA International–Early Development Services (formerly
`Pharma Bio-Research Group B.V, Assen, The Netherlands) using a validated
`high-pressure liquid chromatography-tandem mass spectrometry method with
`two linear concentration ranges. The low calibration range was 1.00 to 100 ng/ml
`for both analytes, using aliquots of 200 ␮l of plasma. The high calibration range
`was 10 to 10,000 ng/ml for both analytes, using aliquots of 100 ␮l of plasma. The
`precision of the low concentration assay ranged from 0.7 to 7.0% for saquinavir
`and from 3.6 to 7.8% for ritonavir. The accuracy ranged from 93.1 to 100.2% and
`from 94.2 to 95.8% for saquinavir and ritonavir, respectively. The precision of the
`
`TABLE 1. Summary of demographic characteristics of the
`study populations
`
`Characteristic
`
`Study arm 1 (n ⫽ 29)
`
`Study arm 2 (n ⫽ 13)
`
`No. of males/no. of
`females
`Black/Caucasian
`Mean age in yr
`(range)
`Mean body wt in
`kg (range)
`Mean body mass index
`in kg/m2 (range)
`
`27/2
`
`1/28
`33.0 (19–62)
`
`13/0
`
`0/13
`30.5 (19–58)
`
`76.7 (45.0–107.8)
`
`80.4 (64.7–100.0)
`
`24.5 (18.0–29.5)
`
`25.1 (20.1–29.2)
`
`high concentration assay ranged from 3.5 to 6.8% for saquinavir and from 4.2 to
`6.1% for ritonavir. The accuracy ranged from 98.9 to 101.4% and from 95.0 to
`103.5% for saquinavir and ritonavir, respectively. Total ketoconazole plasma
`concentrations were analyzed by AAIPharma Deutschland GmbH & Co. KG
`(formerly AAI Deutschland GmbH & Co. KG, Neu-Ulm, Germany) using a
`validated high-pressure liquid chromatography–fluorescence method with a cal-
`ibration range from 25.0 to 2,500 ng/ml, using aliquots of 100 ␮l of plasma. The
`precision ranged from 2.2 to 6.0%, and the accuracy ranged from 97.8 to 101.0%.
`Pharmacokinetic evaluation. Pharmacokinetic parameters were estimated us-
`ing standard noncompartmental methods (Software WinNonlin Professional,
`version 5.2; Pharsight Corp., Mountain View, CA) and actual sampling times.
`The following pharmacokinetic parameters were directly obtained from the ob-
`served concentration-versus-time data: the maximum plasma concentration
`(Cmax), the time to Cmax (Tmax), and the drug concentration at 12 h after
`administration (C12) for saquinavir/ritonavir or at 24 h after administration (C24)
`for ketoconazole. The area under the drug plasma concentration-time curve from
`time zero until 12 h (AUC0-12) for saquinavir/ritonavir or until 24 h (AUC0-24) for
`ketoconazole was calculated by applying the linear trapezoidal rule. The terminal
`elimination half-life (t1/2) was estimated by ln2/kel, where kel is the terminal
`elimination rate constant determined by linear regression of the last four natural
`log-transformed concentration time points with a maximum exclusion of one
`intermediate concentration time point and fitting with an adjusted residuals
`squared value that is ⱖ0.90. The apparent oral plasma clearance at steady-state
`(CLss/F) was estimated by calculating the dose/AUC0-12 for saquinavir and
`ritonavir and dose/AUC0-24 for ketoconazole.
`Statistical analysis. In both study arms, predose concentrations measured on
`the last 2 to 4 days of periods 1 and 2 were summarized per study arm and study
`day. Pharmacokinetic parameters were summarized per study arm and treatment
`for all subjects who completed the trial. For the assessment of the drug-drug
`interaction, the study variables were AUC0-12 and Cmax for saquinavir/ritonavir
`and AUC0-24 and Cmax for ketoconazole. Natural log-transformed values of these
`parameters were used, and the exposure ratios were determined in arm 1 for
`saquinavir and ritonavir for day 28 to day 14 and in arm 2 for ketoconazole for
`day 20 to day 6. The geometric means of the individual exposure ratios, together
`with the corresponding two-sided 90% CIs, were calculated. No formal confir-
`matory hypothesis testing was planned, and P values were interpreted in an
`exploratory manner. The statistical analysis was performed using software SAS
`v8.2 (SAS Institute, Inc., Cary, NC).
`
`RESULTS
`
`Demographics. A total of 42 healthy subjects were enrolled
`in the present study. A total of 29 subjects (27 males and 2
`females) were randomly assigned to study arm 1 assessing
`the effect of ketoconazole on the plasma concentrations of
`saquinavir/ritonavir, and 13 subjects (all males) were randomly
`assigned to study arm 2 assessing the effect of saquinavir/
`ritonavir and the plasma concentrations of ketoconazole. Each
`of the 42 subjects received at least one dose of study drug(s),
`and 32 subjects (20 in arm 1, 12 in arm 2) completed the study
`as planned. The demographic characteristics of the study pop-
`ulation are shown in Table 1.
`
`2
`
`

`

`VOL. 53, 2009
`
`INTERACTION STUDY OF KETOCONAZOLE AND SAQUINAVIR
`
`611
`
`TABLE 2. Geometric mean predose plasma concentrations of saquinavir/ritonavir in study arm 1 (n ⱕ 29)
`
`Mean concn (␮g/ml)a
`
`Agent
`
`Study arm 1, period 1, at day:
`
`Study arm 1, period 2, at day:
`
`Saquinavir
`Ritonavir
`Ketoconazole
`
`11
`
`0.84 (120)
`0.29 (130)
`
`12
`
`0.97 (71)
`0.33 (76)
`
`13
`
`0.99 (74)
`0.38 (40)
`
`14
`
`1.3 (84)
`0.38 (74)
`
`25
`
`0.67 (90)
`0.32 (89)
`0.93 (44)
`
`26
`
`0.84 (120)
`0.30 (84)
`1.0 (38)
`
`27
`
`1.0 (100)
`0.40 (75)
`1.0 (44)
`
`28
`
`1.3 (66)
`0.45 (59)
`1.1 (36)
`
`a The percent CV is indicated in parentheses.
`
`Evaluation of predose concentrations. Serial predose mea-
`surements of the last 2 to 4 days of each treatment period are
`summarized per study arm and treatment day in Tables 2 and
`3. In study arm 1, the daily saquinavir and ritonavir predose
`concentrations were of similar magnitude within each mea-
`sured period, with pronounced but comparable interindividual
`variabilities expressed as the %CV in both periods (Table 2).
`In addition, the daily saquinavir and ritonavir predose concen-
`trations were of similar dimensions in the absence or presence
`of ketoconazole coadministration. Likewise, in arm 2, the daily
`ketoconazole predose concentrations were stable within each
`measured period but were ⬃17-fold higher in the presence of
`saquinavir/ritonavir coadministration compared to ketoco-
`nazole treatment alone (Table 3). Also, in study arm 2 period
`2, the daily saquinavir and ritonavir predose concentrations
`were not dissimilar from those seen for saquinavir/ritonavir
`throughout study arm 1.
`Assessment of drug-drug interactions. The pharmacokinetic
`interaction was evaluated in the 20 subjects in arm 1 and the 12
`subjects in arm 2, who completed the entire study. Figure 1
`shows the 12-h plasma log-transformed concentration-versus-
`time profiles of saquinavir/ritonavir in the absence (day 14) or
`presence of ketoconazole coadministration (day 28) in study
`arm 1, and Fig. 2 shows the respective 24 h profiles of keto-
`conazole in the absence (day 6) or presence of saquinavir/
`ritonavir coadministration (day 20) in study arm 2. Summaries
`of the pharmacokinetic parameters for saquinavir/ritonavir
`(study arm 1) with or without ketoconazole coadministration
`are presented in Table 4, and those for ketoconazole with or
`without saquinavir/ritonavir coadministration are presented in
`Table 5.
`In study arm 1, differences in the plasma concentration-
`versus-time profiles for saquinavir and ritonavir during coad-
`ministration with or without ketoconazole were small and not
`
`TABLE 3. Geometric mean predose plasma concentrations of
`ketoconazole and saquinavir/ritonavir in study arm 2 (n ⱕ 13)
`
`Agent
`
`Study arm 2,
`period 1, at
`day:
`
`Mean concn (␮g/ml)a
`
`Study arm 2, period 2, at day:
`
`5
`
`6
`
`17
`
`18
`
`19
`
`20
`
`Ketoconazole 0.072
`(51)
`
`Saquinavir
`Ritonavir
`
`0.070
`(60)
`
`1.2 (54)
`
`1.3 (54)
`
`1.2 (47)
`
`1.2 (61)
`
`0.83 (62) 1.3 (72)
`1.1 (57)
`1.1 (42)
`0.48 (36) 0.40 (45) 0.42 (70) 0.51 (57)
`
`a The percent CV is indicated in parentheses.
`
`clinically meaningful. Mean values for all pharmacokinetic pa-
`rameters of saquinavir and ritonavir were similar in the ab-
`sence or presence of ketoconazole coadministration, and the
`interindividual variability (as expressed as the %CV) was sim-
`ilar for both compounds and in both of the treatment periods.
`The geometric mean ratio estimates for the AUC0-12 and Cmax
`of saquinavir and ritonavir were close to 1, and all four 90%
`CIs were within the range of 0.86 to 1.26 (Table 4). Based on
`these results, it can be concluded that the addition of ketocona-
`zole at a dose of 200 mg once daily to the approved therapeutic
`regimen of saquinavir/ritonavir at 1,000/100 mg twice daily for
`14 days did not have a clinically relevant effect on the phar-
`macokinetic exposures of saquinavir or ritonavir.
`In study arm 2, the plasma concentration-versus-time pro-
`files showed a considerable increase in ketoconazole exposure
`during coadministration with saquinavir/ritonavir. The termi-
`nal elimination of ketoconazole was prolonged after 14 days of
`coadministration with saquinavir/ritonavir, as indicated by the
`flatter decline in the log-transformed concentration versus
`time profile. The absorption of ketoconazole was minimally
`prolonged during saquinavir/ritonavir coadministration, as ex-
`pressed by a 1-h delay in the median Tmax, from 2.5 h in period
`1 to 3.5 h in period 2. The median t1/2 was also prolonged from
`4.3 h in period 1 to 10.7 h in period 2, and the geometric mean
`
`FIG. 1. Mean plasma concentration-time profiles of saquinavir
`(squares) and ritonavir (circles) after 14 days of saquinavir/ritonavir at
`1,000/100 mg given twice daily (open symbols, dotted line) and after 14
`days of coadministration of saquinavir/ritonavir with ketoconazole at
`200 mg once daily (filled symbols, solid line). The standard deviations
`are shown as error bars.
`
`3
`
`

`

`612
`
`KAESER ET AL.
`
`ANTIMICROB. AGENTS CHEMOTHER.
`
`TABLE 5. Summary of pharmacokinetic parameters for
`ketoconazole in study arm 2 (n ⫽ 12)
`
`Parameter
`
`GM (%CV)a
`
`Period 1
`(ketoconazole
`alone)
`
`Period 2
`(ketoconazole ⫹
`saquinavir ⫹
`ritonavir)
`
`GMR (90% CI)b
`
`3.86 (28.9)
`24.3 (36.4)
`
`5.59 (29.1)
`65.2 (32.1)
`
`1.45 (1.32–1.59)
`2.68 (2.46–2.93)
`
`Cmax (␮g/ml)
`AUC0-24
`(␮g 䡠 h/ml)
`C24 (ng/ml)
`Tmax (h)
`t1/2 (h)
`CLss/F (liter/h)
`a GM, geometric mean.
`b GMR, geometric mean ratio of period 2/period 1. NA, not assessed.
`
`87.4 (79.7)
`2.5 (2.0–5.0)
`4.3 (2.1–7.0)
`8.22 (36.4)
`
`1,100 (47.8)
`3.5 (2.0–4.0)
`10.7 (6.7–18.5)
`3.07 (32.1)
`
`NA
`NA
`NA
`NA
`
`FIG. 2. Mean plasma concentration-time profiles of ketoconazole
`after 6 days of ketoconazole at 200 mg given once daily (open symbols,
`dotted line) and after 14 days of coadministration of ketoconazole with
`saquinavir/ritonavir at 1,000/100 mg twice daily (filled symbols, solid
`line). The standard deviations are shown as error bars.
`
`CLss/F value was decreased by more than 50% with saquinavir/
`ritonavir coadministration, from 8.22 to 3.07 liters/h. Increases
`in the mean C24 values for ketoconazole of ⬎10-fold were seen
`in period 2 relative to period 1. The intersubject variabilities
`(as expressed by the %CV) were comparable in periods 1 and
`2 for ketoconazole. The geometric mean ratio estimates for
`AUC0–24 and Cmax of ketoconazole were substantially larger
`than 1 (Table 5). After 14 days of coadministration of saquina-
`vir/ritonavir, the mean ketoconazole AUC0-24 and Cmax were
`increased by 2.68- and 1.45-fold, respectively, compared to the
`administration of ketoconazole alone.
`Safety results. The study medications were generally well
`tolerated by healthy subjects. There were 10 early discontinu-
`ations from the study: 9 in study arm 1 and 1 in study arm 2.
`The reasons for study withdrawals were adverse events in six
`subjects in arm 1, elevated safety laboratory parameters (high
`triglycerides and low-density lipoprotein cholesterol, respec-
`tively) in two subjects in arm 1, and other reasons in the
`remaining one subject in arm 1 and the one subject in arm 2.
`In study arm 1, the adverse events reporting rates were similar
`in period 1 (days 1 to 14) and period 2 (days 15 to 28), with 62
`
`and 58% of subjects, respectively. Gastrointestinal disorders
`were the most frequently adverse events reported by 41 and
`25% of subjects in periods 1 and 2, respectively, followed by
`events related to infections and infestations, and nervous sys-
`tem disorders reported by 17% of subjects each during sa-
`quinavir/ritonavir treatment in period 1. In study arm 2, the
`reporting rate for adverse events was lower during period 1
`(days 1 to 6) with 23% of subjects compared to that of 92% of
`subjects in period 2 (days 7 to 20). Again, gastrointestinal
`disorders were the most prominent adverse events reported by
`54% of subjects during triple combination therapy in period 2.
`The majority of the adverse events were mild to moderate in
`intensity. All adverse events were resolved without sequelae.
`The greater exposure to ketoconazole when given in combina-
`tion with saquinavir/ritonavir was not associated with unac-
`ceptable safety or tolerability in the present study.
`
`DISCUSSION
`
`In order to assess the impact of drug-drug interaction at the
`CYP3A4 metabolic pathway, strict exclusion criteria were set
`in the present study with regard to concomitant use of
`CYP3A4 substrates, inhibitors, or inducers. Since hormonal
`contraceptives, being CYP3A4 substrates, were not allowed,
`women had to be of non-child-bearing potential or under ef-
`ficient nonhormonal contraceptive protection. With these re-
`quirements, only two women could be recruited into the
`
`TABLE 4. Summary of pharmacokinetic parameters for saquinavir and ritonavir in study arm 1 (n ⫽ 20)a
`
`Saquinavir
`
`GM (%CV)
`
`Ritonavir
`
`GM (%CV)
`
`Period 1
`(⫹ saquinavir)
`
`1.53 (39.4)
`8.9 (36.3)
`0.230 (57.1)
`4.0 (2.0–5.0)
`3.7 (3.1–5.6)
`11.2 (36.3)
`
`Period 2
`(⫹ saquinavir ⫹
`ketoconazole)
`
`1.66 (26.4)
`9.95 (30.3)
`0.292 (57.7)
`4.0 (1.0–5.0)
`4.2 (3.5–5.5)
`8.7 (32.8)
`
`GMR (90% CI)
`
`1.08 (0.96–1.21)
`1.12 (1.03–1.22)
`NA
`NA
`NA
`NA
`
`Parameter
`
`Period 1
`(⫹ ritonavir)
`
`Period 2
`(⫹ ritonavir ⫹
`ketoconazole)
`
`GMR (90% CI)
`
`Cmax (␮g/ml)
`AUC0-12 (␮g 䡠 h/ml)
`C12 (␮g/ml)
`Tmax (h)
`t1/2 (h)
`CLss/F (liter/h)
`a GMR, geometric mean ratio of period 2/period 1. GM, geometric mean. NA, not assessed.
`
`5.01 (51.5)
`30.0 (53.3)
`0.956 (56.3)
`3.0 (2.0–6.0)
`4.9 (4.1–5.9)
`33.4 (53.3)
`
`5.10 (36.3)
`32.2 (40.3)
`1.13 (62.4)
`3.0 (2.0–5.0)
`5.2 (4.5–6.8)
`25.6 (43.3)
`
`1.02 (0.86–1.20)
`1.07 (0.92–1.26)
`NA
`NA
`NA
`NA
`
`4
`
`

`

`VOL. 53, 2009
`
`INTERACTION STUDY OF KETOCONAZOLE AND SAQUINAVIR
`
`613
`
`present study. Both were randomized to study arm 1 and com-
`pleted all study procedures. Knowing that for saquinavir sta-
`tistically significant greater exposures have been observed in
`women than in men (study BP17359 with 87 males and 7
`females [Roche data on file]), the exposures of the two women
`to saquinavir (AUC0-12 ⫽ 47.7 and 43.4 ␮g h/ml in period 1
`and 29.8 and 66.0 ␮g h/ml in period 2) in the present study in
`both periods were similar to those seen in the remaining 18
`men within 40 to 53% variability (Table 4), and therefore the
`conclusions made for the whole population in this study arm
`may also apply to female patients. On the other hand, the
`effect of saquinavir and ritonavir on the exposure of ketoco-
`nazole (arm 2) was only studied in men. However, lacking data
`of female subjects in this study arm, and considering also the
`related safety aspects, it cannot be assumed that this increase
`in ketoconazole exposure during concomitant saquinavir/
`ritonavir at a 1,000/100-mg twice-daily treatment would be of
`different magnitude in women.
`In this study in healthy volunteers, the addition of ketoco-
`nazole at a dose of 200 mg once daily to the approved thera-
`peutic dosing regimen of saquinavir/ritonavir at 1,000/100 mg
`twice daily for 2 weeks did not have a clinically relevant effect
`on the pharmacokinetic exposures of saquinavir and ritonavir.
`For both compounds, the 90% CIs surrounding the geometric
`mean ratio estimates for the AUC0-12 and Cmax were within or
`only 1% exceeding the upper limit of the no-effect boundary
`(0.80 to 1.25) as defined in the U.S. Food and Drug Adminis-
`tration guideline for the industry for in vivo drug interaction
`studies (5).
`By comparison, after 6 days of pretreatment with ketoco-
`nazole at 200 mg once daily, the addition of saquinavir/ritonavir
`at 1,000/100 mg twice daily for 2 weeks increased the exposure
`of ketoconazole by 2.68 (2.46 to 2.93)-fold for AUC0-24 and by
`1.45 (1.32 to 1.59)-fold for Cmax. The median elimination half-
`life for ketoconazole was also more than doubled, from 4.3 h
`after treatment with ketoconazole alone to 10.7 h after coad-
`ministration with saquinavir/ritonavir. The reduced clearance
`of ketoconazole resulted in predose concentrations of ketoco-
`nazole that were ⬎10-fold higher in all subjects after coadmin-
`istration with saquinavir/ritonavir compared to values seen af-
`ter 6 days of treatment with ketoconazole alone.
`The 2-week treatment phases of saquinavir/ritonavir and the
`concomitant triple-drug treatment with ketoconazole was con-
`sidered sufficient in order to achieve stable pharmacokinetic
`conditions for all three medications involved. This treatment
`duration, although longer than required based on the half-lives
`of ritonavir and saquinavir, was selected based on the fact that
`ritonavir not only inhibits the metabolism of CYP3A4 but also
`increases the enzyme activity of CYP3A4 (inhibition-associ-
`ated induction). Due to this autoinduction, plasma concentra-
`tions of saquinavir/ritonavir generally reach steady state 2
`weeks after the start of ritonavir administration (3).
`Studies have already been performed investigating the drug-
`drug interaction between ketoconazole and saquinavir, or
`ritonavir, or saquinavir/ritonavir using several dosing regimens.
`In these studies, the saquinavir and ritonavir doses used were
`different from the approved dosing regimen for saquinavir/
`ritonavir of 1,000/100 mg twice daily. Coadministration of ke-
`toconazole at 400 mg once daily with saquinavir at 1,200 mg
`three times daily increased the AUC of saquinavir by 190%,
`
`but saquinavir did not change the AUC of ketoconazole (see
`the product information for Invirase capsules and tablets
`[Hoffman-La Roche, Inc.]). Coadministration of ritonavir at
`500 mg twice daily with ketoconazole at 200 mg once daily
`resulted in a 3.4-fold increase in the ketoconazole AUC and an
`18% increase in ritonavir AUC (see the product information
`for the Norvir 100-mg capsule [Abbott Laboratories]). Coad-
`ministration of saquinavir/ritonavir at 400/400 mg twice daily
`with ketoconazole at 200 or 400 mg once daily yielded in-
`creases in the saquinavir AUC by 37% and the ritonavir AUC
`by 29% (4). The results of the present study are closest to those
`obtained with the ritonavir and ketoconazole combination, al-
`though the increase in ketoconazole exposure was somewhat
`less for the AUC in the present study (2.68-fold versus 3.4-
`fold) and Cmax (45% in present study versus 55% in the prod-
`uct information for the Norvir 100-mg capsule [Abbott Labo-
`ratories]). The metabolism and elimination of ketoconazole
`are clearly affected by the CYP3A4 inhibitory effect of ritona-
`vir, whereas the influence of ketoconazole, an inhibitor for
`both CYP3A4 and P-glycoprotein, on the saquinavir/ritonavir
`combination is small, as shown by the clinically irrelevant in-
`creases in plasma exposures of these two compounds. In the
`present study, the CYP3A4 inhibitory effect of ketoconazole on
`saquinavir, as seen in the above saquinavir ketoconazole interac-
`tion studies, is, for the most part, superimposed by the CYP3A4
`inhibitory effect of ritonavir on saquinavir. The increase in the
`saquinavir AUC by 37% and the ritonavir AUC by 29% when
`combined with ketoconazole, as observed in an earlier study (4),
`may have been due to the higher ketoconazole dose (400 mg)
`and/or different saquinavir/ritonavir doses (400/400 mg) used
`compared to those in the present study. A full CYP3A inhibitory
`effect of ketoconazole may not have been reached with the keto-
`conazole standard dose of 200 mg daily in the present study.
`However, based on the existing data, and for safety reasons, a
`ketoconazole multiple-dose regimen of 400 mg daily was not
`considered acceptable for this study in healthy volunteers due to
`the hepatic toxicity liability of ketoconazole (see Drug Informa-
`tion Online [http://www.drugs.com/mmx/ketoconazole.html]) and
`as a result of the maximum 200-mg daily ketoconazole dose rec-
`ommendations in combination with ritonavir (product informa-
`tion for the Norvir 100-mg capsule [Abbott Laboratories]). With
`a 400-mg daily dose of ketoconazole in the present study, further
`increases of ketoconazole plasma concentrations would have
`been most likely.
`The safety results as recorded by adverse events and labo-
`ratory safety measurements were consistent with those indi-
`cated in the respective product information for these three
`drugs (see also the product information for the Norvir 100-mg
`capsule [Abbott Laboratories], for Invirase capsules and tab-
`lets [Hoffman-La Roche, Inc.], and for the Nizoral 200-mg
`tablet [Janssen-Cilag, Ltd.]). There were no vital-sign or ECG
`abnormalities of clinical relevance recorded during this study.
`Conclusions. The present two-arm drug-drug interaction
`study involving ketoconazole at 200 mg/day and ritonavir-
`boosted saquinavir (1000 mg of saquinavir and 100 mg of
`ritonavir twice daily) indicated that the Cmax and AUC0-12 of
`both saquinavir and ritonavir are not substantially altered by
`the addition of ketoconazole for the duration of 2 weeks, but
`that, on the other hand, the Cmax and AUC0-24 of ketoconazole
`are increased by 45 and 168%, respectively, after the addition
`
`5
`
`

`

`614
`
`KAESER ET AL.
`
`ANTIMICROB. AGENTS CHEMOTHER.
`
`of saquinavir/ritonavir for 2 weeks. The greater exposure to
`ketoconazole when given in combination with saquinavir/
`ritonavir was not associated with unacceptable safety or toler-
`ability. It is concluded that no dose adjustment for either
`saquinavir or ritonavir is required when coadministered with
`ⱕ200 mg of ketoconazole once daily and, based on the hepa-
`totoxic