- Email: [email protected]
Lack of mutual pharmacokinetic interaction between cerivastatin, a new HMG-CoA reductase inhibitor, and digoxin in healthy normocholesterolemic volunteers
CLINICAL
THERAPEUTICSWOL.
2 1, NO. 9, 1999
Lack of Mutual Pharmacokinetic Interaction Between Cerivastatin, a New HMG-CoA Reductase Inhibitor, and Digoxin in Healthy Normocholesterolemic Volunteers Paul Weber, MD, RPh,‘* John T. Lettieri, PhD,2 Lee Kaiser, PhD,2 and Arthur L. Mazzu, PhD2 ‘Health & SciencesResearch,Inc., Englewood, New Jersey, and ‘The Buyer Corporation, Pharmaceutical Division, WestHaven, Connecticut
ABSTRACT The potential mutual interaction between cerivastatin, a 3-hydroxy-3-methylglutaryLcoenzyme A reductaseinhibitor, and digoxin was assessedin this nonmasked, nonrandomized, multiple-dose study. The effect of cerivastatin 0.2 mg on mean plasma digoxin levels and the effect of digoxin on the single-dose pharmacokinetics of cerivastatin were assessed in 20 healthy normocholesterolemic men between 18 and 45 years of age weighing 140 to 200 lbs (63.3 to 90.0 kg). Subjects were given a single dose of cerivastatin 0.2 mg. After a 2-day washout period, subjects were given a loading dose of digoxin 0.5 mg for 3 days followed by 0.25 mg daily for 5 additional days (period l-digoxin alone). Concurrent dos-
“Current affiliation: ley. New Jersey. Accepted
forpublication
Printed in the USA. Reproduction in whole
0149.1918/99/$19.00
Roche Laboratories,
July
1. 1999.
or part is not permitted
Inc., Nut-
ing with cerivastatin 0.2 mg continued for 14 days (period 2-digoxin and cerivastatin), followed by an g-day course of digoxin-only administration and an optional 6-day extension of digoxin-only treatment for a total of 14 days (period 3). Safety was assessed through physical examination, electrocardiography, laboratory tests, and ophthalmologic examination. Ratio analyses of mean digoxin plasma trough levels, 24-hour urinary digoxin levels, and digoxin clearancewith and without concurrent cerivastatin dosing also were carried out. In addition, single-dose pharmacokinetic variables for cerivastatin, including area under the curve (AUC,-,,), peak concentration (Cmax),time to peak concentration (T,,X), and elimination half-life (t1,2), were examined with and without concurrent digoxin dosing. Eleven of the 20 subjects completed the entire study. Seven subjects discontinued the study because of treatment-emergentadverseevents or laboratory abnormalities that were mostly unrelated to cerivastatin, and 2 subjects were discontinued becauseof protocol vi1563
CLINICAL
olations. Treatment-emergent adverse events developed in 12 subjectsreceiving cerivastatin; 11 of these subjectswere receiving digoxin concurrently. Six adverse events that led to discontinuation of treatment were unrelated to cerivastatin but were related to digoxin or to a preexisting condition. The most commonly reported event was headache,which occurred with equal frequency compared with placebo groups in large cerivastatin clinical trials. Other events were mild or moderate and resolved without intervention. Mild and transient elevations in hepatic transaminase and creatine kinase values (all <2 times the upper limit of normal) were observed in 7 subjects.After 14 days of concurrent dosing of cerivastatin and digoxin. steady-state digoxin plasma levels, urinary digoxin levels, and urinary digoxin clearance were unchangedcompared with steady-state digoxin levels when digoxin was given alone. Compared with dosing with digoxin alone, the AUC,,-,,, Cmax, and t,,, for cerivastatin increased 3%, 20%, and 7%. respectively, while the Tmax was reduced by 18% during concurrent treatment with digoxin. Thesechangesare minimal and would not be expected to be clinically relevant. These results demonstrate that when cerivastatin is administered concurrently with digoxin. neither digoxin nor cerivastatin plasmalevels are altered. The combination therapy wasgenerally well tolerated. Key words.- HMGCoA reductase inhibitor, cerivastatin, statins, hypercholesterolemia, digoxin. INTRODUCTION Congestive heart failure (CHF) affects -1% of the US population (>4 million people), and its incidence continues to increase; after hypertension, hypercholes1564
THERAPEUTICS@
terolemia is the second most frequent causeof CHF.‘-s Furthermore, the prevalence of dyslipidemic hypertension may be high. In 11.200 men and women assessedaspart of a cross-sectionalsurveillance program of coronary heart disease risk factors between 1981 and 1990, the frequencies of dyslipidemia and hypertension were 38.0% and 26.5%. respectively; the conjoint frequency was 15.0%.4 CHF is treated with a variety of agents. Although use of angiotensin-converting enzyme (ACE) inhibitors and beta-adrenergic-receptor blockers may improve survival, a diuretic or digoxin is often prescribed as adjunctive therapy.” In 1 study, retrospective analysis showedthat overall medication use included diuretics (82%), ACE inhibitors (53%), nitrates (49%), digoxin (46%), potassium (40%), acetylsalicylic acid (36%), calcium antagonists (20%). warfarin (17%), beta-blockers (15%), and magnesium(1O%).5 Becauseof digoxin’s narrow therapeutic range, drug-drug interactions may lead to increased digoxin concentration and subsequent toxicity. Drug-drug interactions are known to occur between digoxin and quinidine, amiodarone. and verapamil.6.7However, certain lipid-lowering agents reduce digoxin bioavailability and thus interfere with its desired pharmacologic effect. The bile-acid-binding resin cholestyramine and a related agent, colestipol. interrupt the enterohepatic recycling of digoxin, thereby enhancing its elimination.8,9 The 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductaseinhibitors, also known as statins.are now considered to be first-line therapy for the treatment of hypercholesterolemia.‘O~llMeta-analysis has revealed that statins significantly reduce the incidence of cardiovascular
FL’.WEBER ET AL.
morbidity and mortality in patients with hypercholesterolemia.‘? Because of the prevalence of heart failure and hypercholesterolemia, it becomes important to establish the safety, tolerability, and efficcacy OE Qr;iin~. as colnerqgv ‘m Loa~erjrs receiving digoxin. Cerivastatin is a new, wholly synthetic, nonracemic HMG-CoA reductase inhibitor with lipid-lowering activity in the microgp..s&ng~(l~~Qa~gQ~$. AQ3.bQ.g QD, cerj~~~rarj, reduc~~.~)ow-den.ti&v @oprotein up to 31%. I3 In a pivotal study, c.w>ea%*i~l wa ~2~1 ~dk~akd. -wiTh -adverse-event rates similar to those of placebo am+ LW ; &?J’xPkim iir tTlqmh? transaminases or creatine kinase (CK).i4 The primary objective of this study was to compare the steady-state pharmacokinetics of digoxin, with and without conC~~~Tl~~II~‘~~d,~~iC”crfrdL;71,‘iT,?~~~~~~ &&I m&e v&~ners:. A s.*ia7 d xzk+ec2 tive of this study was to determine the effetis of bigoxin rm -the -single-dose pharmacokinetics of cerivastatin.
Study Design and Treatment Regimens This nonrandomized, nonmasked, mulQle-dose study assessed the potentia’l mutual interaction between cerivastatin and digoxin, comparing the steady-state pharmacokinetics of digoxin with and without concurrent dosing of cerivastatin and, secondarily, the single-dose pharmacokinetits of cerivastatin with and without concurrent dosing of digoxin. The clinical trial protocol was approved by the institutional review board at the single study site. Subjects were screened up to 3 weeks before the start of the study. Screening/pretreatment procedures included a complete med-
ical history, physical examination (including vital signs and blood pressure), chest film, laboratory tests (complete blood count, blood chemistry, and urinalysis), 12-lead electrocardiogram (ECG), and a ~~j~,,r”~~~~~~~~~~~JC
eXtiJJJh2n
On the morning of day -3, eligible subjects again underwent screening laboratory tests, and a physical examination was conducted before enrollment. @A day - 2, a S+X& B .2-m% &LX orf ce5 ~a921511 was atinjstered t13e,sLzUsh cerivastatin pharmacokinetics. After a May Ys&wfl -pi&, MalTeT& -wk?, -iigoxin was initiated with a loading dose of 82 IT-i-gg+?ell- ?a- 3 &y s L%&j~~~ JJeK? then maintained on digoxin 0.25 mg for 27 days (day 4 to day 30). By protocol amendment, digoxin dosing (originally to be 30 days) was extended 6 additional ‘c&j% WI Tixy 5 ‘Lz,-xa3w +iYt -q%irm -UCa ‘I44a,v rsever,v $Teii,* fcx ax pa&eim. Concurrent dosing of cerivastatin 0.2 mg -md
6igOXin
o.25
Tflg
‘DegXI
on
dTy
3
XId
continued through day 22 for a total of 14 days of combination treatment. Five to 7 &.y. t!.om t?x ki.~~.&ATeQ.fdigwh, ahd sampie was drawn from each subject to verify that digoxin was no longer present in the plasma. Digoxin was taken daily at 8 AM, and cerivastatin was taken after each evening meal at --6 PM. The protoco’l design is summarized in the figure. Study Population Male volunteers were between 18 and 45 years of age, weighed 140 to 2W lbs (63.3 to 90.0 kg), and were judged to be healthy on the basis of medical history, physical examination, and laboratory screening. Subjects could not have received any medication for 72 hours before administration of study drug and must 1565
CLINICAL
THERAPEUTICS”
DIG 0.5 mg Maintenance
-CER 0.2 mg +
t -2
dose DIG 0.25 mg
CER 0.2 mg Period
4
e Period
I 1
Period ---*
3
6
8910
20
f DIG only
22
t DIG + CER Treatment
30
34
36
f DIG only
Days
Figure. Protocol design. Primary comparison,period 1 versus period 2; secondary comparison. period 3 versus period 1. DIG = digoxin; CER = cerivastatin. have abstained from alcoholic beverages for 2 days before and throughout the study. All subjects gave written informed consent. Subjects were excluded if they had a history of gastrointestinal disorders,a diagnosisof malignancy or psychiatric disorder, a known hypersensitivity to HMGCoA reductase inhibitors, or a history of alcohol or drug abuse.Other grounds for exclusion were abnormalities in specific laboratory variables, as follows: hematocrit <35%, serum transaminasevalues above normal, serum CK value >2 times the upper limit of normal (ULN) at screening, or total cholesterol < 170 mg/dL. Concomitant medications (including nonsteroidalanti-inflammatory drugs,antacids, and vitamins) were not permitted, with the exception of acetaminophen. which was permitted on an as-neededbasis. Pharmacokinetic Analysis Digoxirl
Plasmadigoxin (12 hours postdose)was measuredfrom blood samplestakenat -8 PM on days4,6 through8.13,16, and20 through 22 in 13 subjects.Seven of these13 subjects 1566
alsohadplasmadigoxin measured at 12hours postdoseon days 34 through 36. Subjects were included in the analysis of digoxin plasmaand urine levels and digoxin clearanceif they haddatafrom at least2 daysduring days6 through8 anddays20 through22. For those subjects electing to continue digoxin dosing until day 36 (per protocol amendment),the plasmadigoxin levels obtainedon days 34 through36 wereanalyzed and consideredsupportive if an interaction had been detectedon days 20 through 22. The rate of return toward the meandigoxin baselinelevels (days 6 through 8) was analyzed by comparingthe meandigoxin levels obtainedon days34 through 36. The on-line plasmadigoxin levels (uncorrected) were normalized using concurrently measuredexternal standards(0.5, 1.5. and 2.5 ng/mL per the manufacturer’s recommendations [Abbott Laboratories, Abbott Park, Illinois]). Digoxin levels were measured using an on-line, open-label method employing an Abbott TDx autoanalyzer. The on-line plasmadigoxin levels (uncorrected) were normalized using a digoxin control value measured concurrently. The corrected value was designated as “digoxin (corrected).” An additional set
P. WEBER ET AL.
of sampleswas split and run both on-line and in batch mode by Medical Research Laboratories,Cincinnati, Ohio, and aredesignated“digoxin (S-3)” For plasmadigoxin assay,the within-run coefficient of variation rangedfrom 5.8% to 1.9% for the lowest reference standardconcentration (0.75 ng/mL) to the highest concentration (3.5 ng/mL). respectively. The between-runcoefficient of variation ranged from 4.3% to 0% acrossthe samereferencedoserange. This study was conducted in a hospital metabolic unit where all urine collections were supervised. Total volume, urinary creatinine concentration, and urinary creatine/total volume were assessedin each subject. Urine was collected on day -2 for 24 hours beginning after the evening dose of cerivastatin. Additional 24-hour urine collections were made on days 6 through 8 and 20 through 22 and were usedto determine urinary digoxin levels and digoxin clearance.For the urine digoxin assay,the within-run coefficient of variation ranged from 8.9% to 3.5% for the lowest reference standard concentration (1.O ng/mL) to the highest concentration (3.0 ng/mL), respectively. Since the urine digoxin assay was done in a single-batch run, no between-run coefficient of variation wascalculated. The digoxin clearance, adjusted for body surface area (BSA), was calculated using the following formula: Digoxin ClCGUlCC’
Dig&n in urine (ng/mL) Digoxin in plasma IlghL)
VOIUIIX of urine x-x-x24 h
1.73 m’ BSA
1h 60min
The “corrected” digoxin concentrations in plasma were used in the calculations above. The digoxin assay is a highly specific immunofluorescenceassay.Although the
cross-reactivity of cerivastatin with the digoxin assay was not formally tested, it hasbeen previously demonstratedthat no cross-reactivity was observed with pravastatin in the digoxin assay.15Therefore, the potential for cross-reactivity of cerivastatin (or metabolites)with the digoxin assay was consideredto be remote. Cerivastutin Subjectswere includedin the analysisof cerivastatin pharmacokineticsif they had complete cerivastatin plasmalevel data on day -2 and day 9. Cerivastatin was measuredin plasmasamplesdrawn immediately before the doseand at 0.5, 1, 1.5, 2, 3,4, 6, 8, 12, 16, and24 hoursafter the doseon day -2 and day 9. Blood sampleswere collected under reducedlighting in tubes containing ethylenediamine-tetraaceticacid. Within 30 minutes after collection, each samplewas centrifuged and the plasmastoredat -2O’C until ready for assayby high-performance liquid chromatography(HPLC) usingfluorescencespectroscopy.16,17 The lower limit of detection was 25 pg/mL. Both interday and intraday precisionwere >90% (coeffrcientsof variation were < 10% at all concentrationstested).Accuracy was characterized by ~10% deviation from nominal concentrations of known standards.The principal variables for cerivastatin pharmacokinetics were areaunder the curve (AUC,,,), peak concentration(C,,), time to peak concentration (T,,), and eliminationhalf-life (t&. Noncompartmentalmethods were used to estimatepharmacokineticvariablesl* Safety Assessment All observations pertinent to the safety of cerivastatin and digoxin were recorded, including laboratory analyses, physical examination findings, adverseevents, and 1567
CLINICAL
vital signs. All subjects were included in the safety analysis.A medical history and a complete physical examination, including measurements of vital signs and supine and standing blood pressure, a chest film, and a 12-lead ECG. were performed at screening. An ophthalmologic examination was performed at screening or at day -2. Subjects had 12-lead ECGs and rhythm strips recorded before and 2 hours after the cerivastatin dose on day -2 and day 9. Additional ECGs were recorded on days 1, 3, 6. 7, 8, 10, 13, 16, 20, 22, 26, 36, and at end-of-study follow-up 5 to 7 days after the last dose of digoxin. On day 23, subjectsunderwent a final ophthalmologic examination. Complete exit physical examinations were performed on day 30 or day 36. Laboratory tests included blood chemistry, thyroid hormone levels, complete blood count, urinalysis, and a screen for human immunodeficiency virus antibody and hepatitis B antigen. Special laboratory testsincluded 24-hour urine collection for measurementof creatinine, a,-microglobulin, and N-acetyl-P-n-glucosaminidase. Statistical Analysis Single-dose pharmacokinetics of cerivastatin were assessedby analyzing the logarithm of day -2 and day 9 AUC,_,,, Cn13.x’ TmaxTand t,,z using analysis of variance (ANOVA) with terms for subject and dosing condition (cerivastatin alone and with concurrent digoxin). This is equivalent to a paired t test. The dosing conditions were compared using a 90% twosided confidence interval (CI) and a two-sided test of the hypothesis of equality of dosing conditions. The logarithm of the geometric mean plasma digoxin level was analyzed using ANOVA with terms 1568
THERAPEUTICS’
for subject and dosing conditions (digoxin alone,digoxin with concurrentcerivastatin, and digoxin after cerivastatin administration). The primary comparisonwasthe geometric meanplasmadigoxin level on days 6, 7, and 8 (digoxin only) relative to the meanplasmalevel of digoxin on days 20, 2 1. and 22 (concurrent digoxin and cerivastatin). The dosing conditions were comparedusinga 90% two-sidedCI and a twosided test of the hypothesisof equality of dosing conditions. Urinary digoxin levels and digoxin clearancewere analyzed in a similar manner. RESULTS Study Population All 20 subjectsenrolled in the study met the enrollment criteria specifiedin the protocol; 13completed the study through day 23 (end of combination treatment), and 11 completed the entire study, including the extended drug dosing. Minor protocol deviations. which were allowed with the concurrence of the study sponsor,occurred in several subjects. Two subjects had total cholesterol levels ~170 mg/dL at the screeningvisit ( 160 and 162 mg/dL). and 1 subject was entered in the study with a day -3 CK value of 39 1 U/L ( 1.7 ULN). Nine subjectswere discontinuedfrom the study prematurely. Seven subjectsdropped out because of adverse events (gynecomastia, abnormalECG [n = 31,tooth pain, positivity for hepatitis B and hepatitis A. and depressionplus chest pain related to heartburn). In addition, I subject was discontinued because of a positive drug screen,and I was discontinuedbecauseof noncompliancewith the study regimen.The 3 subjectswho were discontinuedfrom the study as a result of clinically significant
P. WEBER ET AL.
ECG abnormalities(involving nonspecific ST-segmentand T-wave changes)were in the digoxin-only treatment period (either 0.25 mg or 0.5 mg once daily). These abnormalitieshad resolvedat the time of follow-up. The patient who developed gynecomastiawas taking only digoxin at the time of discontinuation.(Gynecomastiais a known sideeffect of digoxin.) No clinically significantECG changeswereobservedduring concurrent dosing of digoxin and cerivastatin. Although we cannot exclude the possibility that cerivastatin contributed to gynecomastiain this study, the incidence of gynecomastiafrom long-term cerivastatin therapy is rare. In a 6-month pivotal study, 1 of 627 patients receiving cerivastatin (0.05, 0.1, 0.2. and 0.3 mg QD, -150 patients per treatment arm) experienced gynecomastia,similar to the incidence seen with placebo (0 of 154).ij Furthermore, during 2 years of continuous exposure to cerivastatin, the incidence of gynecomastia was 0 of 158 (0.05 mg). 2 of 155 (0.1 mg), 0 of 159 (0.2 mg), and 0 of 155 (0.3 mg), compared with 0 of 153 patients receiving lovastatin 40 mg.i9 Subjects’ mean age was 29.9 years (range, 20-43 years). Their mean height and weight were 179.6 cm (range, 163-191 cm) and 77.9 kg (range, 66-89 kg), respectively. Fifty percent were white, 35% were black, and 15% were Hispanic. All 20 enrolled subjects were given at least the day -2 single dose of 0.2 mg cerivastatin and at least 1 doseof digoxin; thus all were assessable for safety. Eleven subjects completed the entire course of the study; they received 0.2 mg cerivastatin for a total of 15 days and 0.25 mg digoxin for at least 30 days; 4 of these subjects received 0.2 mg cerivastatin for
15 days (day -2 and days 9 through 22) and digoxin for 30 days (days 1 through 30). Seven of these 11 subjectsreceived digoxin for 36 days (optional extension of protocol). Twelve subjectswere valid for all pharmacokinetic analyses,3 were valid only for cerivastatin pharmacokinetic analysis, and 5 were not valid for any pharmacokinetic measurements. Pharmacokinetic Analysis Distribution of Protocol-Valid Patients Cerivastatinpharmacokineticswere measured in 12 of the 20 subjects who were valid for all pharmacokinetic analyses. Three were valid only for cerivastatinpharmacokinetic analysis,and 5 were not valid for any pharmacokineticmeasurements. Influence of Cerivastatin on Digoxin Plasma Levels Of the original 20 patients who participated in this study. 7 were excluded from the analysis.Three subjectswere excluded becauseof ECG abnormalities related to digoxin intolerance before they received cerivastatin plus digoxin. One subject was discontinued and invalidated becauseof a positive illicit drug screen. One subject was discontinued becauseof noncompliance(failure to take scheduledstudy medication while an outpatient). Two additional patients were excluded, I because of an abscessed tooth during the early portion of the study and the other becauseof hepatitis not detected at the initiation of the study. All thesepatients failed to meet the protocol requirement of having dogixin data from at least 2 days within days 6 to 8 and within days 20 to 22. The influence of cerivastatin on plasma digoxin levels, urine digoxin levels, and 1569
CLINICAL
THERAPEUTICS”
Table I. Influence of cerivastatin on plasma and urine concentrations of digoxin and digoxin clearance (n = 13, except as otherwise noted). Geometric Means (ng/mL)
Uncorrected digoxin Corrected digoxin S-3 digoxin* Urine digoxin Clearance
Period I
Period 2 Period 3
0.60 0.59 0.67 94.11 115.91
0.62 0.65 0.68 96.10 109.03
0.54 0.56 0.62 NA NA
Ratio (90’10 CI) Period 1 vs Period 2 Period 3 vs Period 1 1.03 1.10 1.01 1.02 0.94
(0.91-1.17) (1.00-1.20) (0.93-1.09) (0.91-1.14) (0.82-1.08)
0.90 (0.76-I .06) 0.94 (0.84-I .05) 0.91 (0.85-0.99) NA NA
Period 1 = digoxin only: Period 2 = digoxin + cerivastatin; Period 3 = digoxin only (after cerivastatin S-3 = samples split and run both on line and in batch mode: NA = not available, ‘n = 6.
calculated results for digoxin clearance for the 3 discrete study periods are summarized in Table I. Study periods included dosing with digoxin alone before cerivastatin dosing (period I), concurrent digoxin and cerivastatin dosing (period 2) and dosing with digoxin alone after cerivastatin dosing (period 3). Table I summarizesgeometric meandigoxin concentrations expressedas uncorrected. corrected, and S-3 values, as well as urine digoxin levels and calculated digoxin clearance; it also provides a ratio analysis of the meansfor each study period to determine whether cerivastatin influenced digoxin levels, either during concomitant administration or 14 days after cessation of cerivastatin dosing. Plasmalevels of digoxin given concurrently with 0.2 mg cerivastatin did not differ from digoxin plasmalevels following digoxin-only dosing (period 2 vs period 1). The follow-up measurements(period 3) of digoxin levels occurred on days 34 through 36, 14 days after the last dose of the concurrent digoxinkerivastatin regimen (period 2); 7 of the 13 subjectshad additional plasma samples drawn for 1.570
dosing):
digoxin assayon days 34 through 36. The digoxin plasmalevels found in these patients during period 3 were nearly unchanged from those in period 1 (6% to 10% lower in the follow-up period compared with period 1, digoxin-alone dosing); no significant difference in any digoxin plasma level was observed with any method of measurement.In all subjects, digoxin plasma levels were undetectable (or close to undetectable) at the follow-up visit S to 7 days after the last dose of digoxin. Collectively, these data indicate that cerivastatin does not intluence plasma digoxin levels. In addition. there were no clinically relevant changes in urinary digoxin levels or urinary digoxin clearance in the comparison of period 1 and period 2. Influence of D&ox-in 011Cerivastatin Pharmacokinetics Of the 20 patients who received the first dose of cerivastatin on day -2, 15 subjects received the day 9 dose; thus only 15 subjectswere valid for the cerivastatin pharmacokinetic analysis. The 5 subjects excluded from the cerivastatin pharmaco-
P. WEBER ET AL.
kinetic analysis were the first 5 subjects excluded from the digoxin pharmacokinetic analysis. Cerivastatin plasma concentrations were measured using a specific HPLC method in 15 valid subjects on day -2 (single dose of cerivastatin) and on day 9 (first dosing of the combination of cerivastatin and digoxin). The *ucc-,4 values for day -2 and day 9 were 14.19 and 14.67 ng.hr/mL, respectively. Although not a primary variable for this study, the AU& was calculated to be 14.30 and 14.81 ng.hr/mL for day -2 and day 9, respectively. Ratios of AUC,,, and *UC,-_ were 1.034 and 1.035, respectively, and met the requirements for bioequivalence. The Cmaxof cerivastatin showeda slight increasefrom 2.16 ng/mL on day 2 to 2.58 ng/mL on day 9 (ratio of 1.196, 90% CI 1.10-l .30), with a commensuratereduction in the Tmax(from 3.19 to 2.60 hours, ratio of 0.81). The t,,? of cerivastatin was not significantly altered (3.07 vs 3.29 hours, ratio of 1.07). Safety Results Each of the 20 subjects reported 1 or more treatment-emergent adverse events or intercurrent illnessesduring the study (Table II). Headache and abnormal ECG changeswere the mostfrequently reported adverse events. An abnormal ECG was observedin 3 subjectsandresulted in their discontinuation from the study before they received cerivastatin. Most of the treatment-emergent adverse events reported were mild.* eventually resolved, and ei*Mild = transient (<24 hours) and generally not interfering with normal activities; moderate = interfering with normal activities; severe = preventing normal activities.
ther were not related or were remotely related to the study medication. Three reported adverseevents were consideredby the investigator to be of moderate intensity: reported flu (diarrhea and vomiting), tooth decay, and gynecomastia (1 patient each). One patient exhibited suicidal ideation during the study that was judged by the investigator as severe; this event was consideredby a consulting psychiatrist to be unrelated to study drug and determined to be related to a preexisting condition. Twelve (60%) of the 20 subjects reported treatment-emergentadverse events with datesof onset coinciding with cerivastatin treatment (day -2 to day -1, days 9-23). During the study, 5 subjects received acetaminophenas a concomitant medication for analgesicrelief. No clinically significant laboratory abnormalities in either hepatic transaminase (>3 times the ULN) or CK (>5 times the ULN) were reported as treatment-emergent adverse events. One subject with fluctuating minor elevations in alanine aminotransferase (ALT) both during screening and after 19 days of the study was found to test positively for hepatitis A and B antibodies and was thus discontinued from the study on day 11. One patient exhibited elevated aspartateaminotransferase(65 U/L, range O-50) andALT (93 U/L, range O-55) levels on days 16 and 22, respectively. Six subjectsexperienced elevated CK levels, with the highest levels ranging from 244 to 461 (normal range, 0 to 235 U/L). Although 5 of the 7 subjectsdeveloped the highest levels while receiving cerivastatin, the magnitude of the increasein each casewas <2 times the upper limit of normal. Furthermore, the highest valuesin 3 subjectswere isolated and transient; the remaining 2 subjects developed the highest levels af1571
CLINICAL THERAPEUTICS’
Table II. Adverse event summary (N = 20). Adverse Event Headache Rhinitis Electrocardiogram abnormality Infection Diarrhea Asthenia Neck pain Syncope Nausea Tooth pain Vomiting Depression Cough increased Pharyngitis Ear disorder Gynecomastia Rash Abdominal pain Accidental injury Chest pain Flu syndrome Pain Gum hemorrhage Maculopapular rash Amblyopia Conjunctivitis
During Study, no. (5%)
During Cerivastatin. no. (96)
7 (35) 3 (15) 3 (15) 2 (10) 2 (10) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) I (5) 1 (5) 1 (5)
3 (15) 1 (5) 2 (10) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5)
ter discontinuing cerivastatin, that is, beyond day 22. All transaminase and CK elevations returned to within normal ranges in 17 days after peak elevations occurred. No clinically significant ophthalmologic changes were observed. However, 1 subject was found to have a superficial punctate keratopathy at the end of the study that had not been present at study entry. 1572
DISCUSSION
AND CONCLUSIONS
Because of the narrow therapeutic range of digoxin, drug interactions that significantly elevate plasma and hence tissue digoxin levels at steady state may place the cardiac patient at risk. Patients with heart failure and hypercholesterolemia constitute a sizable and growing popula-
P. WEBER
ET AL.
tion, necessitating assessment of potential drug-drug interactions between lipidlcowe~+~~ a_gen’Is anb tiLgoim,Trj,s S~I@Y was designed to assess the potential for change in plasma levels of digoxin when that drug is given concurrently with cerivastatin. The effect of digoxin on the pharmacokinetics of cerivastatin was also assessed. Close monitoring for adverse events, physical examinations, ECGs, ophthalmologic examinations, and labor&3 5ests Sll hT>v+ed shat c~,~ar,~e,T~~~ dosed cerivastatin and digoxin were well tco5eralecj.wXn nomimwarb eYiec1s cSeiu3~ attributed to cerivastatin either alone or in combination (Table II). There were no statistically significant differences in plasma and urine levels of digoxin after 14 days of concurrent dosing of cerivastatin and digoxin (days 20, 21, and 22) compared with initial steadystate digoxin values (days 6, 7, and 8). In fact, all measureddigoxin variables, including plasma and urine levels and digoxin clearance,showedbioequivalence (90% CI, 80% to 125%) between digoxin dosedto steady statewith and without administration of cerivastatin (Table I). Tnese restilts tiemonSrrdre‘in& ceiivastatin doesnot interfere with digoxin pharmacokinetics and that digoxin can be administered safely with cerivastatin withcut coflcera far digoxin accumulation. In addition, when cerivastatin was dosed with digoxin compared with cerivastatin-alone dosing, the AUC0-24 and Cmax, and t,,, for cerivastatin, increas& 3%, 20%,-and 7%, resprstiivel~, while the TmaX decreased 18%. The changesin <&basta& AW&, Tmax,and t,,7 when cerivastatin was concurrently choseh*tin ligoY;rn weye n& slzMi~+~ different from when cerivastatin was gkzn as rn~nu&e~apy. &kh.w& Qk in-
crease in Cmaxwas statistically significant, the magnitude of this increase was ibmos;r\h;jtim’tierm$e ti’timeptin&ence. and it is unlikely to be of clinical importance. Thus the overal pharmacokinetic behavior of cerivastatin is unchangedduring coadministration with digoxin. The magnitude of the changes in cerivastatin pharmacokinetics as a result of cotreatment with digoxin is less than that observed when pravastatin is coadiT2i~TiS~~~~~~ With &@Xiil. Dii&PXiiI iiTcreased the AUC,,, and C,,, of pravaaaim b,~ 2370 anb 2@50,respec~~ve5y. Ym1 these changes were not statistically significant and were not expected to result in altered efficacy compared with pravastatin monotherapy.*O In addition, fluvastatin did not significantly increase the AUC of digoxin.‘O However, for fluvastatin, atorvastatin, and simvastatin, US product labeling recommendsappropriate patient monitoring during combination therapy with digoxin. Digoxin shows no significant interaction with lovastatin (US product labeling). Close monitoring of adverseevents and findings from physical and ophthalmologic exanima’hon,YXYJ. anh ‘1doortiory le3ks showedthat cerivastatin and digoxin, when dosed concurrently, were well tolerated with no untoward effects clearly related to cerivastatin alone or in combinatian. The most commonly reported sideeffects were headacheand rhinitis (Table II). This sideeffect profile of cerivastatin is similar ts5 that of other statins used in cerivastatin clinical tials t0 date.lJ The most inlpuflant adverse event associatedwith statin theraip’1(rn$giaj ‘wasn& ~CFR&&by any patient receiving cerivastatin alone or with ~~~~~nr”l’n~-n~~~~~~~~~-~~~~~~~~$~ associatedwith digoxin (changeson ECG) UGCXXd u-K+-w&3x $xkKts wcxt?,-G-xzti~ 1573
CLINICAL
ing digoxin as monotherapy during initial digoxin treatment. In conclusion, these results suggest that the clinical effectiveness and lack of a drug-drug interaction between cerivastatin and digoxin make it an appropriate agent in the treatment of hypercholesterolemic patients who may also require cotreatment with digoxin.
THERAPEUTICS@
of practice in 4606 acute care patients with congestive heart failure. The relative importance of age. sex, and medical therapy. Arch Izztern Med. 1996:156: 1669-1673. 6. Rodin SM. Johnson BE Pharmacokinetic interactions with digoxin. Clin Phczrwzacokirzet. 1988:15:227-244. 7. Marcus FI. Pharmacokinetic interactions between digoxin and other drugs. J Am Coil Cardiol. 1985:5:82A-90A.
ACKNOWLEDGMENT The authors thank Kenneth B. Pomerantz, PhD, for editorial contributions to this manuscript.
Address correspondence
to: Arthur
Mazzu, PhD, Department of Clinical Pharmacology, Bayer Corporation, Pharmaceutical Division, 400 Morgan Lane, West Haven, CT 06516.
REFERENCES 1. Cheng TO. Congestive heart failure in coronary artery disease. Am J Med. I99 1; 91:409-415. 2. Dhalla NS, Afzal N, Beamish RE. et al. Pathophysiology of cardiac dysfunction in congestive heart failure. Can J Cczrdiol. 1993:9:873X387. 3. Gheorghiade M, Benatar D. Konstam MA, et al. Pharmacotherapy for systolic dysfunction: A review of randomized clinical trials. Am J Ccrrdiol. 199780: 14H-27H. 4. Eaton CB. Feldman HA, Assaf AR. et al. Prevalence of hypertension, dyslipidemia, and dyslipidemic hypertension. J Funz Pmct. 1994;38: 17-23. 5. Clinical Quality Improvement Network Investigators. Mortality risk and patterns
1574
8. Brown DD, Juhl RP, Warner SL. Effects of hypocholesterolemic interventions on digoxin bioavailability. Drzzg Metah Rev. 1979:9:107-l 17. 9. Brown DD. Juhl RP, Warner SL. Decreased bioavailability of digoxin due to hypocholesterolemic interventions. Circu(ation. 1978:58:164-172. 10. Prevention of coronary heart disease in clinical practice. Recommendations of the Second Joint Task Force of European and Other Societies on Coronary Prevention. Eur Heart J. 1998:19:1434-1503. 11. Gonzalez ER. Preventing cardiovascular atlerosclerosis: Role of HMG-CoA reductase inhibitors. Fomzzdazy 1996;3 1:582602. 12. Hebert PR. Gaziano JM. Chan KS, Hennekens CH. Cholesterol lowering with statin drugs, risk of stroke, and total mortality. An overview of randomized trials. JAMA. 1997;278:313-321. 13. Stein E. Sprecher DL. Allenby KS, et al. Cerivastatin. a new potent synthetic HMG-CoA reductase inhibitor: Effect of 0.2 mg/day in subjects with primary hypercholesterolemia. J Curdio\sasc Plzarmacol Ther. 1997:2:7-16. 14. Insull W. Stein E. Whalen E. Ripa S, and the Cerivastatin Study Group. Cerivastatin: A new potent HMG-CoA reductase
P. WEBER ET AL.
inhibitor: Efficacy and tolerability in primary hypercholesterolemia. J Am Co11 Curdiol. 1997:29:46A. 15. Triscari J, Swanson BN, Willard DA, et al. Steady state serum concentrations of pravastatin and digoxin when given in combination. Br J Clin Pharmacol. 1993;36:263-265. 16. Krol GJ, Beck GW, Ritter W, Lettieri JT. LC separation and induced fluorometric detection of rivastatin in blood plasma. J Pharm Biomed Anal. 1993;11:1269-1275. 17. Krol GJ, Beck GW, Ritter W, et al. Determination of rivastatin levels in plasma
samples by HPLC and enzyme inhibition assays. In: Reid R. Wilson ID, eds. Biqfluid and Tissue Atlalysis for Drugs, Including Hypolipidaemics. Vol 23. Cambridge: Royal Society of Chemistry: 1994:147-156. 18. Gibaldi M, Perrier D. Pharmucokinerics. New York: Marcel Dekker: 1980:293-300. 19. Data on file, Study X3 l-091. West Haven, Conn: Bayer Corporation; May 1996. 20. Deslypere JP. Clinical implications of the biopharmaceutical properties of fluvastatin. Am J Cardiol. 1994;73: 12D-17D.
1575
THERAPEUTICSWOL.
2 1, NO. 9, 1999
Lack of Mutual Pharmacokinetic Interaction Between Cerivastatin, a New HMG-CoA Reductase Inhibitor, and Digoxin in Healthy Normocholesterolemic Volunteers Paul Weber, MD, RPh,‘* John T. Lettieri, PhD,2 Lee Kaiser, PhD,2 and Arthur L. Mazzu, PhD2 ‘Health & SciencesResearch,Inc., Englewood, New Jersey, and ‘The Buyer Corporation, Pharmaceutical Division, WestHaven, Connecticut
ABSTRACT The potential mutual interaction between cerivastatin, a 3-hydroxy-3-methylglutaryLcoenzyme A reductaseinhibitor, and digoxin was assessedin this nonmasked, nonrandomized, multiple-dose study. The effect of cerivastatin 0.2 mg on mean plasma digoxin levels and the effect of digoxin on the single-dose pharmacokinetics of cerivastatin were assessed in 20 healthy normocholesterolemic men between 18 and 45 years of age weighing 140 to 200 lbs (63.3 to 90.0 kg). Subjects were given a single dose of cerivastatin 0.2 mg. After a 2-day washout period, subjects were given a loading dose of digoxin 0.5 mg for 3 days followed by 0.25 mg daily for 5 additional days (period l-digoxin alone). Concurrent dos-
“Current affiliation: ley. New Jersey. Accepted
forpublication
Printed in the USA. Reproduction in whole
0149.1918/99/$19.00
Roche Laboratories,
July
1. 1999.
or part is not permitted
Inc., Nut-
ing with cerivastatin 0.2 mg continued for 14 days (period 2-digoxin and cerivastatin), followed by an g-day course of digoxin-only administration and an optional 6-day extension of digoxin-only treatment for a total of 14 days (period 3). Safety was assessed through physical examination, electrocardiography, laboratory tests, and ophthalmologic examination. Ratio analyses of mean digoxin plasma trough levels, 24-hour urinary digoxin levels, and digoxin clearancewith and without concurrent cerivastatin dosing also were carried out. In addition, single-dose pharmacokinetic variables for cerivastatin, including area under the curve (AUC,-,,), peak concentration (Cmax),time to peak concentration (T,,X), and elimination half-life (t1,2), were examined with and without concurrent digoxin dosing. Eleven of the 20 subjects completed the entire study. Seven subjects discontinued the study because of treatment-emergentadverseevents or laboratory abnormalities that were mostly unrelated to cerivastatin, and 2 subjects were discontinued becauseof protocol vi1563
CLINICAL
olations. Treatment-emergent adverse events developed in 12 subjectsreceiving cerivastatin; 11 of these subjectswere receiving digoxin concurrently. Six adverse events that led to discontinuation of treatment were unrelated to cerivastatin but were related to digoxin or to a preexisting condition. The most commonly reported event was headache,which occurred with equal frequency compared with placebo groups in large cerivastatin clinical trials. Other events were mild or moderate and resolved without intervention. Mild and transient elevations in hepatic transaminase and creatine kinase values (all <2 times the upper limit of normal) were observed in 7 subjects.After 14 days of concurrent dosing of cerivastatin and digoxin. steady-state digoxin plasma levels, urinary digoxin levels, and urinary digoxin clearance were unchangedcompared with steady-state digoxin levels when digoxin was given alone. Compared with dosing with digoxin alone, the AUC,,-,,, Cmax, and t,,, for cerivastatin increased 3%, 20%, and 7%. respectively, while the Tmax was reduced by 18% during concurrent treatment with digoxin. Thesechangesare minimal and would not be expected to be clinically relevant. These results demonstrate that when cerivastatin is administered concurrently with digoxin. neither digoxin nor cerivastatin plasmalevels are altered. The combination therapy wasgenerally well tolerated. Key words.- HMGCoA reductase inhibitor, cerivastatin, statins, hypercholesterolemia, digoxin. INTRODUCTION Congestive heart failure (CHF) affects -1% of the US population (>4 million people), and its incidence continues to increase; after hypertension, hypercholes1564
THERAPEUTICS@
terolemia is the second most frequent causeof CHF.‘-s Furthermore, the prevalence of dyslipidemic hypertension may be high. In 11.200 men and women assessedaspart of a cross-sectionalsurveillance program of coronary heart disease risk factors between 1981 and 1990, the frequencies of dyslipidemia and hypertension were 38.0% and 26.5%. respectively; the conjoint frequency was 15.0%.4 CHF is treated with a variety of agents. Although use of angiotensin-converting enzyme (ACE) inhibitors and beta-adrenergic-receptor blockers may improve survival, a diuretic or digoxin is often prescribed as adjunctive therapy.” In 1 study, retrospective analysis showedthat overall medication use included diuretics (82%), ACE inhibitors (53%), nitrates (49%), digoxin (46%), potassium (40%), acetylsalicylic acid (36%), calcium antagonists (20%). warfarin (17%), beta-blockers (15%), and magnesium(1O%).5 Becauseof digoxin’s narrow therapeutic range, drug-drug interactions may lead to increased digoxin concentration and subsequent toxicity. Drug-drug interactions are known to occur between digoxin and quinidine, amiodarone. and verapamil.6.7However, certain lipid-lowering agents reduce digoxin bioavailability and thus interfere with its desired pharmacologic effect. The bile-acid-binding resin cholestyramine and a related agent, colestipol. interrupt the enterohepatic recycling of digoxin, thereby enhancing its elimination.8,9 The 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductaseinhibitors, also known as statins.are now considered to be first-line therapy for the treatment of hypercholesterolemia.‘O~llMeta-analysis has revealed that statins significantly reduce the incidence of cardiovascular
FL’.WEBER ET AL.
morbidity and mortality in patients with hypercholesterolemia.‘? Because of the prevalence of heart failure and hypercholesterolemia, it becomes important to establish the safety, tolerability, and efficcacy OE Qr;iin~. as colnerqgv ‘m Loa~erjrs receiving digoxin. Cerivastatin is a new, wholly synthetic, nonracemic HMG-CoA reductase inhibitor with lipid-lowering activity in the microgp..s&ng~(l~~Qa~gQ~$. AQ3.bQ.g QD, cerj~~~rarj, reduc~~.~)ow-den.ti&v @oprotein up to 31%. I3 In a pivotal study, c.w>ea%*i~l wa ~2~1 ~dk~akd. -wiTh -adverse-event rates similar to those of placebo am+ LW ; &?J’xPkim iir tTlqmh? transaminases or creatine kinase (CK).i4 The primary objective of this study was to compare the steady-state pharmacokinetics of digoxin, with and without conC~~~Tl~~II~‘~~d,~~iC”crfrdL;71,‘iT,?~~~~~~ &&I m&e v&~ners:. A s.*ia7 d xzk+ec2 tive of this study was to determine the effetis of bigoxin rm -the -single-dose pharmacokinetics of cerivastatin.
Study Design and Treatment Regimens This nonrandomized, nonmasked, mulQle-dose study assessed the potentia’l mutual interaction between cerivastatin and digoxin, comparing the steady-state pharmacokinetics of digoxin with and without concurrent dosing of cerivastatin and, secondarily, the single-dose pharmacokinetits of cerivastatin with and without concurrent dosing of digoxin. The clinical trial protocol was approved by the institutional review board at the single study site. Subjects were screened up to 3 weeks before the start of the study. Screening/pretreatment procedures included a complete med-
ical history, physical examination (including vital signs and blood pressure), chest film, laboratory tests (complete blood count, blood chemistry, and urinalysis), 12-lead electrocardiogram (ECG), and a ~~j~,,r”~~~~~~~~~~~JC
eXtiJJJh2n
On the morning of day -3, eligible subjects again underwent screening laboratory tests, and a physical examination was conducted before enrollment. @A day - 2, a S+X& B .2-m% &LX orf ce5 ~a921511 was atinjstered t13e,sLzUsh cerivastatin pharmacokinetics. After a May Ys&wfl -pi&, MalTeT& -wk?, -iigoxin was initiated with a loading dose of 82 IT-i-gg+?ell- ?a- 3 &y s L%&j~~~ JJeK? then maintained on digoxin 0.25 mg for 27 days (day 4 to day 30). By protocol amendment, digoxin dosing (originally to be 30 days) was extended 6 additional ‘c&j% WI Tixy 5 ‘Lz,-xa3w +iYt -q%irm -UCa ‘I44a,v rsever,v $Teii,* fcx ax pa&eim. Concurrent dosing of cerivastatin 0.2 mg -md
6igOXin
o.25
Tflg
‘DegXI
on
dTy
3
XId
continued through day 22 for a total of 14 days of combination treatment. Five to 7 &.y. t!.om t?x ki.~~.&ATeQ.fdigwh, ahd sampie was drawn from each subject to verify that digoxin was no longer present in the plasma. Digoxin was taken daily at 8 AM, and cerivastatin was taken after each evening meal at --6 PM. The protoco’l design is summarized in the figure. Study Population Male volunteers were between 18 and 45 years of age, weighed 140 to 2W lbs (63.3 to 90.0 kg), and were judged to be healthy on the basis of medical history, physical examination, and laboratory screening. Subjects could not have received any medication for 72 hours before administration of study drug and must 1565
CLINICAL
THERAPEUTICS”
DIG 0.5 mg Maintenance
-CER 0.2 mg +
t -2
dose DIG 0.25 mg
CER 0.2 mg Period
4
e Period
I 1
Period ---*
3
6
8910
20
f DIG only
22
t DIG + CER Treatment
30
34
36
f DIG only
Days
Figure. Protocol design. Primary comparison,period 1 versus period 2; secondary comparison. period 3 versus period 1. DIG = digoxin; CER = cerivastatin. have abstained from alcoholic beverages for 2 days before and throughout the study. All subjects gave written informed consent. Subjects were excluded if they had a history of gastrointestinal disorders,a diagnosisof malignancy or psychiatric disorder, a known hypersensitivity to HMGCoA reductase inhibitors, or a history of alcohol or drug abuse.Other grounds for exclusion were abnormalities in specific laboratory variables, as follows: hematocrit <35%, serum transaminasevalues above normal, serum CK value >2 times the upper limit of normal (ULN) at screening, or total cholesterol < 170 mg/dL. Concomitant medications (including nonsteroidalanti-inflammatory drugs,antacids, and vitamins) were not permitted, with the exception of acetaminophen. which was permitted on an as-neededbasis. Pharmacokinetic Analysis Digoxirl
Plasmadigoxin (12 hours postdose)was measuredfrom blood samplestakenat -8 PM on days4,6 through8.13,16, and20 through 22 in 13 subjects.Seven of these13 subjects 1566
alsohadplasmadigoxin measured at 12hours postdoseon days 34 through 36. Subjects were included in the analysis of digoxin plasmaand urine levels and digoxin clearanceif they haddatafrom at least2 daysduring days6 through8 anddays20 through22. For those subjects electing to continue digoxin dosing until day 36 (per protocol amendment),the plasmadigoxin levels obtainedon days 34 through36 wereanalyzed and consideredsupportive if an interaction had been detectedon days 20 through 22. The rate of return toward the meandigoxin baselinelevels (days 6 through 8) was analyzed by comparingthe meandigoxin levels obtainedon days34 through 36. The on-line plasmadigoxin levels (uncorrected) were normalized using concurrently measuredexternal standards(0.5, 1.5. and 2.5 ng/mL per the manufacturer’s recommendations [Abbott Laboratories, Abbott Park, Illinois]). Digoxin levels were measured using an on-line, open-label method employing an Abbott TDx autoanalyzer. The on-line plasmadigoxin levels (uncorrected) were normalized using a digoxin control value measured concurrently. The corrected value was designated as “digoxin (corrected).” An additional set
P. WEBER ET AL.
of sampleswas split and run both on-line and in batch mode by Medical Research Laboratories,Cincinnati, Ohio, and aredesignated“digoxin (S-3)” For plasmadigoxin assay,the within-run coefficient of variation rangedfrom 5.8% to 1.9% for the lowest reference standardconcentration (0.75 ng/mL) to the highest concentration (3.5 ng/mL). respectively. The between-runcoefficient of variation ranged from 4.3% to 0% acrossthe samereferencedoserange. This study was conducted in a hospital metabolic unit where all urine collections were supervised. Total volume, urinary creatinine concentration, and urinary creatine/total volume were assessedin each subject. Urine was collected on day -2 for 24 hours beginning after the evening dose of cerivastatin. Additional 24-hour urine collections were made on days 6 through 8 and 20 through 22 and were usedto determine urinary digoxin levels and digoxin clearance.For the urine digoxin assay,the within-run coefficient of variation ranged from 8.9% to 3.5% for the lowest reference standard concentration (1.O ng/mL) to the highest concentration (3.0 ng/mL), respectively. Since the urine digoxin assay was done in a single-batch run, no between-run coefficient of variation wascalculated. The digoxin clearance, adjusted for body surface area (BSA), was calculated using the following formula: Digoxin ClCGUlCC’
Dig&n in urine (ng/mL) Digoxin in plasma IlghL)
VOIUIIX of urine x-x-x24 h
1.73 m’ BSA
1h 60min
The “corrected” digoxin concentrations in plasma were used in the calculations above. The digoxin assay is a highly specific immunofluorescenceassay.Although the
cross-reactivity of cerivastatin with the digoxin assay was not formally tested, it hasbeen previously demonstratedthat no cross-reactivity was observed with pravastatin in the digoxin assay.15Therefore, the potential for cross-reactivity of cerivastatin (or metabolites)with the digoxin assay was consideredto be remote. Cerivastutin Subjectswere includedin the analysisof cerivastatin pharmacokineticsif they had complete cerivastatin plasmalevel data on day -2 and day 9. Cerivastatin was measuredin plasmasamplesdrawn immediately before the doseand at 0.5, 1, 1.5, 2, 3,4, 6, 8, 12, 16, and24 hoursafter the doseon day -2 and day 9. Blood sampleswere collected under reducedlighting in tubes containing ethylenediamine-tetraaceticacid. Within 30 minutes after collection, each samplewas centrifuged and the plasmastoredat -2O’C until ready for assayby high-performance liquid chromatography(HPLC) usingfluorescencespectroscopy.16,17 The lower limit of detection was 25 pg/mL. Both interday and intraday precisionwere >90% (coeffrcientsof variation were < 10% at all concentrationstested).Accuracy was characterized by ~10% deviation from nominal concentrations of known standards.The principal variables for cerivastatin pharmacokinetics were areaunder the curve (AUC,,,), peak concentration(C,,), time to peak concentration (T,,), and eliminationhalf-life (t&. Noncompartmentalmethods were used to estimatepharmacokineticvariablesl* Safety Assessment All observations pertinent to the safety of cerivastatin and digoxin were recorded, including laboratory analyses, physical examination findings, adverseevents, and 1567
CLINICAL
vital signs. All subjects were included in the safety analysis.A medical history and a complete physical examination, including measurements of vital signs and supine and standing blood pressure, a chest film, and a 12-lead ECG. were performed at screening. An ophthalmologic examination was performed at screening or at day -2. Subjects had 12-lead ECGs and rhythm strips recorded before and 2 hours after the cerivastatin dose on day -2 and day 9. Additional ECGs were recorded on days 1, 3, 6. 7, 8, 10, 13, 16, 20, 22, 26, 36, and at end-of-study follow-up 5 to 7 days after the last dose of digoxin. On day 23, subjectsunderwent a final ophthalmologic examination. Complete exit physical examinations were performed on day 30 or day 36. Laboratory tests included blood chemistry, thyroid hormone levels, complete blood count, urinalysis, and a screen for human immunodeficiency virus antibody and hepatitis B antigen. Special laboratory testsincluded 24-hour urine collection for measurementof creatinine, a,-microglobulin, and N-acetyl-P-n-glucosaminidase. Statistical Analysis Single-dose pharmacokinetics of cerivastatin were assessedby analyzing the logarithm of day -2 and day 9 AUC,_,,, Cn13.x’ TmaxTand t,,z using analysis of variance (ANOVA) with terms for subject and dosing condition (cerivastatin alone and with concurrent digoxin). This is equivalent to a paired t test. The dosing conditions were compared using a 90% twosided confidence interval (CI) and a two-sided test of the hypothesis of equality of dosing conditions. The logarithm of the geometric mean plasma digoxin level was analyzed using ANOVA with terms 1568
THERAPEUTICS’
for subject and dosing conditions (digoxin alone,digoxin with concurrentcerivastatin, and digoxin after cerivastatin administration). The primary comparisonwasthe geometric meanplasmadigoxin level on days 6, 7, and 8 (digoxin only) relative to the meanplasmalevel of digoxin on days 20, 2 1. and 22 (concurrent digoxin and cerivastatin). The dosing conditions were comparedusinga 90% two-sidedCI and a twosided test of the hypothesisof equality of dosing conditions. Urinary digoxin levels and digoxin clearancewere analyzed in a similar manner. RESULTS Study Population All 20 subjectsenrolled in the study met the enrollment criteria specifiedin the protocol; 13completed the study through day 23 (end of combination treatment), and 11 completed the entire study, including the extended drug dosing. Minor protocol deviations. which were allowed with the concurrence of the study sponsor,occurred in several subjects. Two subjects had total cholesterol levels ~170 mg/dL at the screeningvisit ( 160 and 162 mg/dL). and 1 subject was entered in the study with a day -3 CK value of 39 1 U/L ( 1.7 ULN). Nine subjectswere discontinuedfrom the study prematurely. Seven subjectsdropped out because of adverse events (gynecomastia, abnormalECG [n = 31,tooth pain, positivity for hepatitis B and hepatitis A. and depressionplus chest pain related to heartburn). In addition, I subject was discontinued because of a positive drug screen,and I was discontinuedbecauseof noncompliancewith the study regimen.The 3 subjectswho were discontinuedfrom the study as a result of clinically significant
P. WEBER ET AL.
ECG abnormalities(involving nonspecific ST-segmentand T-wave changes)were in the digoxin-only treatment period (either 0.25 mg or 0.5 mg once daily). These abnormalitieshad resolvedat the time of follow-up. The patient who developed gynecomastiawas taking only digoxin at the time of discontinuation.(Gynecomastiais a known sideeffect of digoxin.) No clinically significantECG changeswereobservedduring concurrent dosing of digoxin and cerivastatin. Although we cannot exclude the possibility that cerivastatin contributed to gynecomastiain this study, the incidence of gynecomastiafrom long-term cerivastatin therapy is rare. In a 6-month pivotal study, 1 of 627 patients receiving cerivastatin (0.05, 0.1, 0.2. and 0.3 mg QD, -150 patients per treatment arm) experienced gynecomastia,similar to the incidence seen with placebo (0 of 154).ij Furthermore, during 2 years of continuous exposure to cerivastatin, the incidence of gynecomastia was 0 of 158 (0.05 mg). 2 of 155 (0.1 mg), 0 of 159 (0.2 mg), and 0 of 155 (0.3 mg), compared with 0 of 153 patients receiving lovastatin 40 mg.i9 Subjects’ mean age was 29.9 years (range, 20-43 years). Their mean height and weight were 179.6 cm (range, 163-191 cm) and 77.9 kg (range, 66-89 kg), respectively. Fifty percent were white, 35% were black, and 15% were Hispanic. All 20 enrolled subjects were given at least the day -2 single dose of 0.2 mg cerivastatin and at least 1 doseof digoxin; thus all were assessable for safety. Eleven subjects completed the entire course of the study; they received 0.2 mg cerivastatin for a total of 15 days and 0.25 mg digoxin for at least 30 days; 4 of these subjects received 0.2 mg cerivastatin for
15 days (day -2 and days 9 through 22) and digoxin for 30 days (days 1 through 30). Seven of these 11 subjectsreceived digoxin for 36 days (optional extension of protocol). Twelve subjectswere valid for all pharmacokinetic analyses,3 were valid only for cerivastatin pharmacokinetic analysis, and 5 were not valid for any pharmacokinetic measurements. Pharmacokinetic Analysis Distribution of Protocol-Valid Patients Cerivastatinpharmacokineticswere measured in 12 of the 20 subjects who were valid for all pharmacokinetic analyses. Three were valid only for cerivastatinpharmacokinetic analysis,and 5 were not valid for any pharmacokineticmeasurements. Influence of Cerivastatin on Digoxin Plasma Levels Of the original 20 patients who participated in this study. 7 were excluded from the analysis.Three subjectswere excluded becauseof ECG abnormalities related to digoxin intolerance before they received cerivastatin plus digoxin. One subject was discontinued and invalidated becauseof a positive illicit drug screen. One subject was discontinued becauseof noncompliance(failure to take scheduledstudy medication while an outpatient). Two additional patients were excluded, I because of an abscessed tooth during the early portion of the study and the other becauseof hepatitis not detected at the initiation of the study. All thesepatients failed to meet the protocol requirement of having dogixin data from at least 2 days within days 6 to 8 and within days 20 to 22. The influence of cerivastatin on plasma digoxin levels, urine digoxin levels, and 1569
CLINICAL
THERAPEUTICS”
Table I. Influence of cerivastatin on plasma and urine concentrations of digoxin and digoxin clearance (n = 13, except as otherwise noted). Geometric Means (ng/mL)
Uncorrected digoxin Corrected digoxin S-3 digoxin* Urine digoxin Clearance
Period I
Period 2 Period 3
0.60 0.59 0.67 94.11 115.91
0.62 0.65 0.68 96.10 109.03
0.54 0.56 0.62 NA NA
Ratio (90’10 CI) Period 1 vs Period 2 Period 3 vs Period 1 1.03 1.10 1.01 1.02 0.94
(0.91-1.17) (1.00-1.20) (0.93-1.09) (0.91-1.14) (0.82-1.08)
0.90 (0.76-I .06) 0.94 (0.84-I .05) 0.91 (0.85-0.99) NA NA
Period 1 = digoxin only: Period 2 = digoxin + cerivastatin; Period 3 = digoxin only (after cerivastatin S-3 = samples split and run both on line and in batch mode: NA = not available, ‘n = 6.
calculated results for digoxin clearance for the 3 discrete study periods are summarized in Table I. Study periods included dosing with digoxin alone before cerivastatin dosing (period I), concurrent digoxin and cerivastatin dosing (period 2) and dosing with digoxin alone after cerivastatin dosing (period 3). Table I summarizesgeometric meandigoxin concentrations expressedas uncorrected. corrected, and S-3 values, as well as urine digoxin levels and calculated digoxin clearance; it also provides a ratio analysis of the meansfor each study period to determine whether cerivastatin influenced digoxin levels, either during concomitant administration or 14 days after cessation of cerivastatin dosing. Plasmalevels of digoxin given concurrently with 0.2 mg cerivastatin did not differ from digoxin plasmalevels following digoxin-only dosing (period 2 vs period 1). The follow-up measurements(period 3) of digoxin levels occurred on days 34 through 36, 14 days after the last dose of the concurrent digoxinkerivastatin regimen (period 2); 7 of the 13 subjectshad additional plasma samples drawn for 1.570
dosing):
digoxin assayon days 34 through 36. The digoxin plasmalevels found in these patients during period 3 were nearly unchanged from those in period 1 (6% to 10% lower in the follow-up period compared with period 1, digoxin-alone dosing); no significant difference in any digoxin plasma level was observed with any method of measurement.In all subjects, digoxin plasma levels were undetectable (or close to undetectable) at the follow-up visit S to 7 days after the last dose of digoxin. Collectively, these data indicate that cerivastatin does not intluence plasma digoxin levels. In addition. there were no clinically relevant changes in urinary digoxin levels or urinary digoxin clearance in the comparison of period 1 and period 2. Influence of D&ox-in 011Cerivastatin Pharmacokinetics Of the 20 patients who received the first dose of cerivastatin on day -2, 15 subjects received the day 9 dose; thus only 15 subjectswere valid for the cerivastatin pharmacokinetic analysis. The 5 subjects excluded from the cerivastatin pharmaco-
P. WEBER ET AL.
kinetic analysis were the first 5 subjects excluded from the digoxin pharmacokinetic analysis. Cerivastatin plasma concentrations were measured using a specific HPLC method in 15 valid subjects on day -2 (single dose of cerivastatin) and on day 9 (first dosing of the combination of cerivastatin and digoxin). The *ucc-,4 values for day -2 and day 9 were 14.19 and 14.67 ng.hr/mL, respectively. Although not a primary variable for this study, the AU& was calculated to be 14.30 and 14.81 ng.hr/mL for day -2 and day 9, respectively. Ratios of AUC,,, and *UC,-_ were 1.034 and 1.035, respectively, and met the requirements for bioequivalence. The Cmaxof cerivastatin showeda slight increasefrom 2.16 ng/mL on day 2 to 2.58 ng/mL on day 9 (ratio of 1.196, 90% CI 1.10-l .30), with a commensuratereduction in the Tmax(from 3.19 to 2.60 hours, ratio of 0.81). The t,,? of cerivastatin was not significantly altered (3.07 vs 3.29 hours, ratio of 1.07). Safety Results Each of the 20 subjects reported 1 or more treatment-emergent adverse events or intercurrent illnessesduring the study (Table II). Headache and abnormal ECG changeswere the mostfrequently reported adverse events. An abnormal ECG was observedin 3 subjectsandresulted in their discontinuation from the study before they received cerivastatin. Most of the treatment-emergent adverse events reported were mild.* eventually resolved, and ei*Mild = transient (<24 hours) and generally not interfering with normal activities; moderate = interfering with normal activities; severe = preventing normal activities.
ther were not related or were remotely related to the study medication. Three reported adverseevents were consideredby the investigator to be of moderate intensity: reported flu (diarrhea and vomiting), tooth decay, and gynecomastia (1 patient each). One patient exhibited suicidal ideation during the study that was judged by the investigator as severe; this event was consideredby a consulting psychiatrist to be unrelated to study drug and determined to be related to a preexisting condition. Twelve (60%) of the 20 subjects reported treatment-emergentadverse events with datesof onset coinciding with cerivastatin treatment (day -2 to day -1, days 9-23). During the study, 5 subjects received acetaminophenas a concomitant medication for analgesicrelief. No clinically significant laboratory abnormalities in either hepatic transaminase (>3 times the ULN) or CK (>5 times the ULN) were reported as treatment-emergent adverse events. One subject with fluctuating minor elevations in alanine aminotransferase (ALT) both during screening and after 19 days of the study was found to test positively for hepatitis A and B antibodies and was thus discontinued from the study on day 11. One patient exhibited elevated aspartateaminotransferase(65 U/L, range O-50) andALT (93 U/L, range O-55) levels on days 16 and 22, respectively. Six subjectsexperienced elevated CK levels, with the highest levels ranging from 244 to 461 (normal range, 0 to 235 U/L). Although 5 of the 7 subjectsdeveloped the highest levels while receiving cerivastatin, the magnitude of the increasein each casewas <2 times the upper limit of normal. Furthermore, the highest valuesin 3 subjectswere isolated and transient; the remaining 2 subjects developed the highest levels af1571
CLINICAL THERAPEUTICS’
Table II. Adverse event summary (N = 20). Adverse Event Headache Rhinitis Electrocardiogram abnormality Infection Diarrhea Asthenia Neck pain Syncope Nausea Tooth pain Vomiting Depression Cough increased Pharyngitis Ear disorder Gynecomastia Rash Abdominal pain Accidental injury Chest pain Flu syndrome Pain Gum hemorrhage Maculopapular rash Amblyopia Conjunctivitis
During Study, no. (5%)
During Cerivastatin. no. (96)
7 (35) 3 (15) 3 (15) 2 (10) 2 (10) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) I (5) 1 (5) 1 (5)
3 (15) 1 (5) 2 (10) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5) 1 (5)
ter discontinuing cerivastatin, that is, beyond day 22. All transaminase and CK elevations returned to within normal ranges in 17 days after peak elevations occurred. No clinically significant ophthalmologic changes were observed. However, 1 subject was found to have a superficial punctate keratopathy at the end of the study that had not been present at study entry. 1572
DISCUSSION
AND CONCLUSIONS
Because of the narrow therapeutic range of digoxin, drug interactions that significantly elevate plasma and hence tissue digoxin levels at steady state may place the cardiac patient at risk. Patients with heart failure and hypercholesterolemia constitute a sizable and growing popula-
P. WEBER
ET AL.
tion, necessitating assessment of potential drug-drug interactions between lipidlcowe~+~~ a_gen’Is anb tiLgoim,Trj,s S~I@Y was designed to assess the potential for change in plasma levels of digoxin when that drug is given concurrently with cerivastatin. The effect of digoxin on the pharmacokinetics of cerivastatin was also assessed. Close monitoring for adverse events, physical examinations, ECGs, ophthalmologic examinations, and labor&3 5ests Sll hT>v+ed shat c~,~ar,~e,T~~~ dosed cerivastatin and digoxin were well tco5eralecj.wXn nomimwarb eYiec1s cSeiu3~ attributed to cerivastatin either alone or in combination (Table II). There were no statistically significant differences in plasma and urine levels of digoxin after 14 days of concurrent dosing of cerivastatin and digoxin (days 20, 21, and 22) compared with initial steadystate digoxin values (days 6, 7, and 8). In fact, all measureddigoxin variables, including plasma and urine levels and digoxin clearance,showedbioequivalence (90% CI, 80% to 125%) between digoxin dosedto steady statewith and without administration of cerivastatin (Table I). Tnese restilts tiemonSrrdre‘in& ceiivastatin doesnot interfere with digoxin pharmacokinetics and that digoxin can be administered safely with cerivastatin withcut coflcera far digoxin accumulation. In addition, when cerivastatin was dosed with digoxin compared with cerivastatin-alone dosing, the AUC0-24 and Cmax, and t,,, for cerivastatin, increas& 3%, 20%,-and 7%, resprstiivel~, while the TmaX decreased 18%. The changesin <&basta& AW&, Tmax,and t,,7 when cerivastatin was concurrently choseh*tin ligoY;rn weye n& slzMi~+~ different from when cerivastatin was gkzn as rn~nu&e~apy. &kh.w& Qk in-
crease in Cmaxwas statistically significant, the magnitude of this increase was ibmos;r\h;jtim’tierm$e ti’timeptin&ence. and it is unlikely to be of clinical importance. Thus the overal pharmacokinetic behavior of cerivastatin is unchangedduring coadministration with digoxin. The magnitude of the changes in cerivastatin pharmacokinetics as a result of cotreatment with digoxin is less than that observed when pravastatin is coadiT2i~TiS~~~~~~ With &@Xiil. Dii&PXiiI iiTcreased the AUC,,, and C,,, of pravaaaim b,~ 2370 anb 2@50,respec~~ve5y. Ym1 these changes were not statistically significant and were not expected to result in altered efficacy compared with pravastatin monotherapy.*O In addition, fluvastatin did not significantly increase the AUC of digoxin.‘O However, for fluvastatin, atorvastatin, and simvastatin, US product labeling recommendsappropriate patient monitoring during combination therapy with digoxin. Digoxin shows no significant interaction with lovastatin (US product labeling). Close monitoring of adverseevents and findings from physical and ophthalmologic exanima’hon,YXYJ. anh ‘1doortiory le3ks showedthat cerivastatin and digoxin, when dosed concurrently, were well tolerated with no untoward effects clearly related to cerivastatin alone or in combinatian. The most commonly reported sideeffects were headacheand rhinitis (Table II). This sideeffect profile of cerivastatin is similar ts5 that of other statins used in cerivastatin clinical tials t0 date.lJ The most inlpuflant adverse event associatedwith statin theraip’1(rn$giaj ‘wasn& ~CFR&&by any patient receiving cerivastatin alone or with ~~~~~nr”l’n~-n~~~~~~~~~-~~~~~~~~$~ associatedwith digoxin (changeson ECG) UGCXXd u-K+-w&3x $xkKts wcxt?,-G-xzti~ 1573
CLINICAL
ing digoxin as monotherapy during initial digoxin treatment. In conclusion, these results suggest that the clinical effectiveness and lack of a drug-drug interaction between cerivastatin and digoxin make it an appropriate agent in the treatment of hypercholesterolemic patients who may also require cotreatment with digoxin.
THERAPEUTICS@
of practice in 4606 acute care patients with congestive heart failure. The relative importance of age. sex, and medical therapy. Arch Izztern Med. 1996:156: 1669-1673. 6. Rodin SM. Johnson BE Pharmacokinetic interactions with digoxin. Clin Phczrwzacokirzet. 1988:15:227-244. 7. Marcus FI. Pharmacokinetic interactions between digoxin and other drugs. J Am Coil Cardiol. 1985:5:82A-90A.
ACKNOWLEDGMENT The authors thank Kenneth B. Pomerantz, PhD, for editorial contributions to this manuscript.
Address correspondence
to: Arthur
Mazzu, PhD, Department of Clinical Pharmacology, Bayer Corporation, Pharmaceutical Division, 400 Morgan Lane, West Haven, CT 06516.
REFERENCES 1. Cheng TO. Congestive heart failure in coronary artery disease. Am J Med. I99 1; 91:409-415. 2. Dhalla NS, Afzal N, Beamish RE. et al. Pathophysiology of cardiac dysfunction in congestive heart failure. Can J Cczrdiol. 1993:9:873X387. 3. Gheorghiade M, Benatar D. Konstam MA, et al. Pharmacotherapy for systolic dysfunction: A review of randomized clinical trials. Am J Ccrrdiol. 199780: 14H-27H. 4. Eaton CB. Feldman HA, Assaf AR. et al. Prevalence of hypertension, dyslipidemia, and dyslipidemic hypertension. J Funz Pmct. 1994;38: 17-23. 5. Clinical Quality Improvement Network Investigators. Mortality risk and patterns
1574
8. Brown DD, Juhl RP, Warner SL. Effects of hypocholesterolemic interventions on digoxin bioavailability. Drzzg Metah Rev. 1979:9:107-l 17. 9. Brown DD. Juhl RP, Warner SL. Decreased bioavailability of digoxin due to hypocholesterolemic interventions. Circu(ation. 1978:58:164-172. 10. Prevention of coronary heart disease in clinical practice. Recommendations of the Second Joint Task Force of European and Other Societies on Coronary Prevention. Eur Heart J. 1998:19:1434-1503. 11. Gonzalez ER. Preventing cardiovascular atlerosclerosis: Role of HMG-CoA reductase inhibitors. Fomzzdazy 1996;3 1:582602. 12. Hebert PR. Gaziano JM. Chan KS, Hennekens CH. Cholesterol lowering with statin drugs, risk of stroke, and total mortality. An overview of randomized trials. JAMA. 1997;278:313-321. 13. Stein E. Sprecher DL. Allenby KS, et al. Cerivastatin. a new potent synthetic HMG-CoA reductase inhibitor: Effect of 0.2 mg/day in subjects with primary hypercholesterolemia. J Curdio\sasc Plzarmacol Ther. 1997:2:7-16. 14. Insull W. Stein E. Whalen E. Ripa S, and the Cerivastatin Study Group. Cerivastatin: A new potent HMG-CoA reductase
P. WEBER ET AL.
inhibitor: Efficacy and tolerability in primary hypercholesterolemia. J Am Co11 Curdiol. 1997:29:46A. 15. Triscari J, Swanson BN, Willard DA, et al. Steady state serum concentrations of pravastatin and digoxin when given in combination. Br J Clin Pharmacol. 1993;36:263-265. 16. Krol GJ, Beck GW, Ritter W, Lettieri JT. LC separation and induced fluorometric detection of rivastatin in blood plasma. J Pharm Biomed Anal. 1993;11:1269-1275. 17. Krol GJ, Beck GW, Ritter W, et al. Determination of rivastatin levels in plasma
samples by HPLC and enzyme inhibition assays. In: Reid R. Wilson ID, eds. Biqfluid and Tissue Atlalysis for Drugs, Including Hypolipidaemics. Vol 23. Cambridge: Royal Society of Chemistry: 1994:147-156. 18. Gibaldi M, Perrier D. Pharmucokinerics. New York: Marcel Dekker: 1980:293-300. 19. Data on file, Study X3 l-091. West Haven, Conn: Bayer Corporation; May 1996. 20. Deslypere JP. Clinical implications of the biopharmaceutical properties of fluvastatin. Am J Cardiol. 1994;73: 12D-17D.
1575