Review
`
`No Increased Risk of Ketoconazole Toxicity
`in Drug–Drug Interaction Studies
`
`The Journal of Clinical Pharmacology
`2016, 56(10) 1203–1211
`C(cid:2) 2016, The American College of
`Clinical Pharmacology
`DOI: 10.1002/jcph.795
`
`No´emi Outeiro, MDent, Nicolas Hohmann, MD, and Gerd Mikus, MD, MSc
`
`Abstract
`In July 2013 the U.S. Food and Drug Administration (FDA) released a safety announcement regarding the use of ketoconazole and its adverse drug
`reactions. The FDA report advised against the use ketoconazole tablets as a first-line treatment for any fungal infections because of the risk of
`potentially serious drug–drug interactions and liver and adrenal gland complications. The European Medicines Agency (EMA) also proposed to limit
`the use of oral ketoconazole in fungal infections because of the same risk of harmful effects and interactions. In addition, the FDA also advised against
`the use of oral ketoconazole in drug interaction studies, in which it has been extensively used as an index inhibitor of drug metabolism. The aim of this
`investigation was to evaluate the risks of ketoconazole-induced hepatotoxicity described by the FDA and EMA in published drug interaction studies
`with ketoconazole and compare these data with the toxicity reported for ketoconazole when used as antifungal treatment. In the drug interaction
`studies (2355 participants; healthy volunteers and patients; median treatment duration, 6 days), only 40 participants were reported to have increased
`liver transaminase activity (1.7%), and no deaths were reported or associated with ketoconazole. In studies investigating ketoconazole treatment,
`patients were treated for 276 days (median), and 5.6% of patients had elevated liver enzyme activity. Because of the short treatment period in drug
`interaction studies the risk of drug-induced hepatic injury is considered very low. As such, we recommend that ketoconazole remain a safe CYP3A
`index inhibitor for use in drug interaction studies with healthy volunteers.
`
`Keywords
`CYP3A inhibition, ketoconazole, hepatic toxicity, adverse events
`
`infections im-
`invasive fungal
`The management of
`proved significantly with the introduction of ketocona-
`zole in 1981, as did tolerability and adverse effects,
`compared with amphotericin B.1,2 Triggered by the
`observation of the clinically significant terfenadine–
`ketoconazole drug interaction and after the introduc-
`tion of fluconazole and itraconazole, the clinical utility
`of ketoconazole as an systemic antifungal treatment
`changed. Ketoconazole became widely used as a proto-
`typic CYP3A inhibitor in vitro and in vivo, especially
`to establish the “worst-case” scenario for CYP3A-
`mediated interactions.1,2
`The first reports of adverse drug reactions (ADRs)
`for the broad-spectrum antimycotic drug ketoconazole
`occurred in 1982, reported in the United States and
`other countries.3 The liver was the organ most affected
`by ketoconazole exposure. The effect of
`treatment
`with ketoconazole ranged from asymptomatic transient
`irregularities of liver enzyme activity to potentially fatal
`acute hepatic necrosis. Serious ADRs are usually rare.4
`The recent European Medicines Agency (EMA)–
`approved Summary of Product Characteristics5 for
`ketoconazole lists the following adverse effects (and
`frequency):
`
`Very common (ࣙ1/10): liver function tests abnormal,
`hepatic enzyme increased.
`
`Common (ࣙ1/100 to <1/10): adrenal insufficiency, nau-
`sea, abdominal pain, vomiting, diarrhea, pruritus,
`rash.
`Uncommon: (ࣙ1/1,000 to <1/100): thrombocytopenia,
`allergic conditions including anaphylactic shock, ana-
`phylactoid reaction and anaphylactic reaction, and
`angioedema, headache, dizziness, somnolence, ur-
`ticaria, alopecia, asthenia, platelet count decreased.
`
`In July 2013 the U.S. Food and Drug Administration
`(FDA) released a safety announcement regarding ad-
`verse reactions and drug-induced liver injury (DILI) re-
`lated to ketoconazole exposure.6 In its report the FDA
`advised avoiding the use of ketoconazole tablets as a
`first-line treatment for any fungal infections because
`of the risk of clinically serious drug–drug interactions
`and adverse effects related to hepatic and adrenal
`gland complications. The EMA also recommended a
`
`Department of Clinical Pharmacology and Pharmacoepidemiology,
`University of Heidelberg, Heidelberg, Germany
`
`Submitted for publication 20 May 2016; accepted 8 July 2016.
`
`Corresponding Author:
`Gerd Mikus, MD, Department Clinical Pharmacology and Pharmacoepi-
`demiology, University Hospital Heidelberg, Im Neuenheimer Feld 410,
`69120 Heidelberg, Germany
`Email: gerd.mikus@med.uni-heidelberg.de
`
`1
`
`TEVA1074
`Teva Pharmaceuticals USA, Inc. v. Corcept Therapeutics, Inc.
`PGR2019-00048
`
`

`

`1204
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`restriction on the use of oral ketoconazole in the
`management of fungal infections because of the risk
`of ADRs such as drug interactions, liver toxicity, and
`adrenal insufficiency.7,8
`Moreover, by the end of 2013 the FDA also rec-
`ommended against the use of ketoconazole in drug–
`drug interaction studies,9 although in the current draft
`guideline on “Drug Interaction Studies—Study Design,
`Data Analysis, Implications for Dosing, and Labeling
`Recommendations,”10 the use of ketoconazole as an
`inhibitor of CYP3A is still recommended. A recently
`published analysis of 53 drug–drug interaction (DDI)
`studies (2005–2015) including 971 healthy volunteers
`with systemic ketoconazole exposure reported no en-
`hanced risk of ketoconazole-related toxicity.11 Further-
`more, there is an ongoing discussion concerning the
`optimal alternative to replace ketoconazole as an index
`inhibitor for CYP3A-mediated metabolism for use in
`DDI studies.12,13
`The aim of the present investigation was to eval-
`uate the risks of ketoconazole-induced hepatotoxicity
`as identified by the FDA and EMA in 2013 in all
`published DDI studies conducted using ketoconazole
`and compare these data with the incidence of toxicity
`reported during antifungal ketoconazole treatment.
`
`Methods
`A literature search was carried out by screening
`PUBMED for the terms ketoconazole (“ketocona-
`zole”[MeSH Terms] OR “ketoconazole”[All Fields]).
`Found were 8129 articles; randomized clinical trials,
`longitudinal studies, case reports, and literature reviews
`were included until September 2015.
`All titles and abstracts from the 8129 articles were
`read and divided into several groups accordingly:
`
`r “In vivo” interactions with ketoconazole
`Case reports
`Original research studies
`Literature reviews
`r Toxicity related to ketoconazole use
`Case reports
`Original research studies
`Literature reviews
`r Ketoconazole pharmacokinetic studies
`
`All available full-text articles were analyzed accord-
`ing to study population (healthy volunteers or patients),
`number of participants per study, the substrate used
`(and its dose), and the ketoconazole dose and length
`of treatment period. The total dose of ketoconazole
`was calculated and the number of
`treatment days
`with a 400-mg standard daily dose. In addition, area
`under the plasma concentration–time curve (AUC),
`peak plasma concentration (Cmax), clearance (CL), and
`
`Figure 1. Summary of the total ketoconazole studies.
`
`terminal elimination half-life (t1/2) of the substrate
`with and without ketoconazole coadministration were
`recorded; safety reports and laboratory values indica-
`tive of liver injury including alanine aminotransferase
`(ALT) and aspartate aminotransferase (AST) activity
`were extracted when recorded. Cmax and AUC ratios
`of the substrates were calculated as measures of expo-
`sure during ketoconazole coadministration divided by
`exposure without ketoconazole. Descriptive statistical
`parameters were obtained using Prism 6.0 (GraphPad
`Software, La Jolla, California).
`
`Results
`From the 8129 articles found, 358 reported were
`selected and divided into the groups accordingly
`(Figure 1). Of these 224 full-text articles were acquired.
`The articles used are listed in the supplement.
`
`DDI Studies
`In the 126 articles of the group “in vivo” ketocona-
`zole interaction studies, AUC values extracted from
`120 articles were evaluated. A total of 161 studies
`were analyzed because there was more than 1 study
`reported per article (several substrates tested or differ-
`ent ketoconazole doses compared). From these studies
`106 used a single dose of the substrate, 55 studies
`employed a multiple dose, and 27 studies used both
`single- and multiple-dose regimens in the study design.
`In 10 studies a single oral dose of ketoconazole was
`used, whereas in 151 studies a multiple-dose regimen of
`ketoconazole was employed. From 9 studies calculation
`of AUC ratios (AUCRs) of the “victim” drug was not
`possible, in 6 studies no AUC values for the victim drug
`
`2
`
`

`

`1205
`
`values throughout the study (without providing any
`specific data), 3 explicitly reported no events of hep-
`atic toxicity,15–17 1 article stated there were no ADRs
`attributable to ketoconazole,17 and 18 articles reported
`changes in the laboratory values of ALT and AST
`activity; however, only 5 provided data on ALT and/or
`AST activity of study participants (Table 1).
`The mean total dose of ketoconazole administered
`per study was 2767 mg (median, 2400 mg). The highest
`total cumulative dose of ketoconazole administered
`to healthy volunteers was 11.2 g (400 mg/day for
`28 days), and no serious ADRs were reported in this
`study.18 The highest total cumulative dose administered
`to patients with advanced prostate carcinoma was
`16.8 g (400 mg 3 times daily for 14 days), and there
`were no reported episodes of hepatotoxicity, although
`fatigue and weakness were reported, and it the severity
`was reported to increase with higher ketoconazole
`doses.16
`Three studies in healthy volunteers reported an
`episode of increased transaminase activity. In 1 study
`15 healthy volunteers were administered ketoconazole
`200 mg twice daily for 9 days, and 1 participant with-
`drew from the study because of increased transami-
`nase activity19; in another study 42 healthy volunteers
`received ketoconazole 400 mg once daily for 9 days,
`and 2 participants withdrew prematurely because of
`elevated hepatic enzyme activity and headache.20 One
`participant was reported to have abnormal laboratory
`tests including elevated liver function tests after keto-
`conazole administration and before taking the study
`drug (L-771,688), which was considered unrelated to
`the study. The participant was found to be infected with
`the hepatitis C virus.14
`In a clinical trial involving patients 1 participant of
`13 cancer patients receiving 400 mg ketoconazole/day
`for 4 days was reported to have elevated total biliru-
`bin serum concentrations without a history of liver
`disease.21
`To account for different ketoconazole dosing regi-
`mens the treatment duration of a standardized dose of
`400 mg/day of ketoconazole was calculated. Healthy
`participants were administered ketoconazole for a me-
`dian of 6 days, with 5 days the most often used
`administration duration, followed by 7 days. For the
`drug interaction studies involving patients the median
`duration of ketoconazole treatment was 5 days, with
`3 days the most common period of ketoconazole ad-
`ministration.
`
`Toxicity During Therapy Studies With Ketoconazole
`Fifty-five articles that reported studies in which ke-
`toconazole was used as an antifungal treatment were
`analyzed—36 of these article were clinical studies, 12
`were literature reviews, and 7 were case reports. In some
`
`Drug interaction studies
`
`healthy
`patients
`
`20
`
`40
`
`50
`
`Toxicity studies
`
`>1500
`
`Outeiro et al
`
`400
`
`200
`
`0
`
`800
`
`600
`
`400
`
`200
`
`Number of study participants
`
`40
`
`50
`
`1500
`
`0
`
`20
`
`500
`1000
`Treatment duration (days)
`Figure 2. Treatment duration of ketoconazole treatment (normalized
`to 400 mg/day) in drug interaction and treatment studies.
`
`were provided, and in 3 studies incomplete values (only
`ketoconazole AUC or only the victim drug’s AUC) were
`provided. The majority of DDI studies were performed
`in healthy volunteers (82.5%), and only 17.5% were
`carried out in patients with the target indication or renal
`impairment. In total, 2355 participants were exposed to
`ketoconazole in the drug interaction studies captured in
`this analysis.
`The duration of ketoconazole administration varied
`considerably in the 151 drug interaction studies, shown
`in Figure 2 (range, 2–29 days; median, 6 days). In
`most cases ketoconazole was administered for 5 days;
`the second most common duration of administration
`used in these studies was 7 days. The administered
`doses of ketoconazole varied among 400 mg once daily
`(79 studies), 400 mg twice daily (1 study), 400 mg 3
`times daily (1 study), 200 mg once daily (40 studies),
`200 mg twice daily (34 studies), 200 mg 3 times daily (1
`study), 100 mg once daily (2 studies), 100 mg twice daily
`(1 study), and 600 mg once daily (2 studies). In some of
`the studies more than 1 dose of ketoconazole was tested
`(eg, 400 mg once daily and 200 mg once daily).
`From the 161 interaction studies, 152 reported
`AUC values, and the AUCR of the victim drug with
`and without ketoconazole coadministration was calcu-
`lated. The highest AUCR was approximately 60.0 for
`L-771,688, a strong and highly selective experimental
`α1-adrenoceptor antagonist with demonstrated effects
`on clinical urodynamics.14 All drugs with an AUCR
`above 2.0 (moderate to strong inhibition) are presented
`in Figure 3.
`
`Toxicity Reported in Interaction Studies
`Of the 126 articles reporting drug interaction studies,
`20 did not report or describe any ADRs, and 3 articles
`reported ADRs without adequate details of the event,
`leaving 103 articles available for analysis. From the
`103 articles, 32 did not report any details of liver
`function tests, 49 reported no changes in the laboratory
`
`3
`
`

`

`1206
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`1
`
`2
`
`5
`
`AUCR
`10
`
`20
`
`50
`
`L-771,688
`ebastine
`nisoldipine
`ibrutinib
`voclosporin
`ABT-384
`everolimus
`midazolam
`triazolam
`casopitant
`ridaforolimus
`bosutinib
`budesonide
`midazolam
`quetiapine
`midazolam
`midostaurin
`tolvaptan
`levo-acetyl-a-methadol
`neratinib
`ranitidine
`maraviroc
`mirodenafil
`sotrastaurin
`loratadine
`casopitant
`alprazolam
`lapatinib
`ranolazine
`domperidone
`ranolazine
`udenafil
`crizotinib
`nilotinib
`saquinavir
`tamsulosin
`drospirenone
`rivaroxaban
`tolterodine
`saxagliptin
`oxycodone
`darunavir
`ixabepilone
`methylprednisolone
`antipyrine
`drospirenone
`ixabepilone
`bosentan
`ixabepilone
`tolterodine
`axitinib
`cinacalcet
`paritaprevir
`apixaban
`
`Figure 3. AUC ratios (AUCRs) of the victim drug with and without ketoconazole extracted from the screened literature; only AUCRs greater than
`2 (moderate to strong inhibition) are shown.
`
`cases there was more than 1 study per article (different
`periods of administration or different ketoconazole
`doses), providing a total of 80 studies included in this
`analysis. The total number of patients treated with ke-
`toconazole was 4427, and the duration of ketoconazole
`treatment and/or the dose given was provided in 72
`studies.
`The duration of ketoconazole treatment was highly
`variable (see Figure 2). The median ketoconazole treat-
`ment duration was 276 days, and the median total
`cumulative dose of ketoconazole administered was
`73 g.
`In 1 study ketoconazole was reported to be used
`for a period of 3650 days (10 years, 51 patients)
`coadministered with cyclosporine in kidney transplant
`patients.22 In this study the ketoconazole dose regimen
`
`was 100 mg/day, and in some patients the daily dose
`was reduced to 50 mg/day. This study includes a control
`group that did not receive ketoconazole treatment.
`The total ketoconazole dose administered was typically
`365 g (when taken as 100 mg daily).
`The adverse effects observed and reported in this
`study were:
`
`r Abnormal liver function tests in 25.5% of the
`ketoconazole-treated group (n = 51) and 30.6%
`of the control group (n = 49);
`r Transient elevation of hepatic transaminases,
`which was associated with high blood
`concentrations
`of
`cyclosporine
`and
`normalized with dose adjustment (4 patients
`in the ketoconazole-treated group);
`
`4
`
`

`

`Outeiro et al
`
`1207
`
`Table 1. All Studies With Reported ALT and AST Values in Ketoconazole Drug Interaction Studies
`
`Author, Year
`
`Number of Participants
`
`ALT (U/L)
`
`AST (U/L)
`
`Engels, Mathot, Loos, van
`Schaik, Verweij, 200638
`
`7 (cancer patients)
`
`Xu et al, 201539
`
`22 (healthy volunteers)
`
`Kovarik, Huang, Slade, Sfikas,
`Chandler, 200940
`
`17 (healthy volunteers)
`
`Lee et al, 199541
`
`13 (cancer patients)
`
`Tham et al, 200642
`
`29 (cancer patients)
`
`Baseline
`20 U/L
`No mention during or after the study
`1 subject ࣘ 144 U/L
`Decreased once ketoconazole was
`stopped
`1 subject 74 U/L (normal range, 3–60
`U/L)
`2 months after the study, 18 U/L
`
`Baseline
`25 U/L
`No mention during or after the study
`–
`
`–
`
`1 subject 479 IU/mL
`On day 20 of treatment with
`ketoconazole
`Average data
`Control, 21 U/L
`Ketoconazole, 29 U/L
`
`1 subject 535 IU/mL
`On day 20 of treatment with
`ketoconazole
`Average data
`Control 33 U/L
`Ketoconazole 34 U/L
`
`r Persistent mildly increased bilirubin concentra-
`tions and serum transaminase activity (n = 2
`patients);
`r 5 Patients were withdrawn from therapy be-
`cause of viral hepatitis;
`r 16 Patients discontinued ketoconazole ther-
`apy at different times (6–72 months) for rea-
`sons including hepatic dysfunction (n = 5),
`antituberculostatic drugs (n = 4), pregnancy
`(n = 5), noncompliance with immunosuppres-
`sive medication (n = 1), and development of
`cancer (n = 1).22
`
`The highest cumulative ketoconazole dose admin-
`istered to patients with hormone-refractory prostate
`cancer was 1296 g over a period of 1440 days.23 In
`this study 38 patients received ketoconazole 300 mg
`3 times daily together with 30 mg of hydrocorti-
`sone. The duration of treatment varied between 3 and
`48 months, and the median was 6 months. The following
`adverse effects were reported: nausea (13.2%), fatigue
`(10.6%), diarrhea (2.6%), visual disturbance (2.6%),
`and abnormal liver function tests (2.6%; ALT and AST
`activity values not provided), and 6 patients (15.8%)
`were reported to have discontinued ketoconazole ther-
`apy because of “intolerable adverse effects.”23 At the
`end of the study 12 patients were reported to survive
`(the median age was approximately 73 years).
`Allergic reactions after a single oral dose of ke-
`toconazole have been reported.24 A female patient
`with dermatomycosis of the breasts was prescribed
`ketoconazole, and according to the patient, she had
`not previously experienced an allergic reaction except
`for contact dermatitis to miconazole. Shortly (45 min-
`utes) after administration of ketoconazole, the patient
`reported being dizzy and collapsed several
`times.
`
`During the examination severe angioedema and pro-
`nounced erythema of the face, hands, and feet were
`found. After sufficient supportive treatment all symp-
`toms resolved the next day.24 A male patient also
`received a single dose of ketoconazole (200 mg). He
`also had not previously experienced an allergic re-
`action to ketoconazole and reported not currently
`taking other medication. After 30 minutes the patient
`reported swelling of the hands and feet and itching.
`Clinical examination revealed severe angioedema of the
`face, hands, and feet. Gastrointestinal complaints and
`pronounced dyspnea were also present but gradually
`resolved. After supportive treatment the patient recov-
`ered within 2 hours.24
`From the 55 articles analyzed, only 4 actually re-
`ported ALT and AST values, 4 articles reported ALT
`values only (Table 2), 18 articles described abnor-
`mal liver function tests without providing any values,
`11 studies reported no changes in hepatic function, and
`18 articles did not mention hepatic function.
`Two articles reported ALT and AST activity at the
`beginning of ketoconazole treatment and reported no
`changes throughout the 2 years of follow-up.25,26
`
`Discussion
`According to a report from the FDA, ketoconazole
`should not be used in drug interaction studies because
`the typical dosage (200–400 mg daily for 5 days) has
`been associated with liver injury.9 The decision of
`the FDA was based on data from the FDA Adverse
`Event Reporting System (FAERS), which is not easily
`accessible to the public, and epidemiological data of
`the General Practice Research Database from 1999,27 in
`which 2 of 1052 patients with prescribed ketoconazole
`reported drug-induced liver injury (nonfatal outcome).
`
`5
`
`

`

`1208
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`Table 2. All Studies With Reported ALT and AST Values in Patients During Ketoconazole Treatment (Only Average Values)
`
`Author, Year
`
`Lewis et al, 19843
`
`Study Design
`
`Number of Participants
`
`ALT (U/L)
`
`AST (U/L)
`
`Literature review (33
`cases)
`
`33 (patients)
`
`Range, 70 to >2000 Average, 613
`Duration of treatment 1–69 days
`
`Range 50 to >2000
`Average, 594
`Duration of treatment,
`1–69 days
`
`–
`
`–
`
`At the start of the
`study
`Ketoconazole, 19.20
`Control, 20.29
`–
`
`Ketoconazole, 21.31
`Control, 21.14
`–
`
`Average data, 21
`
`Sobh et al, 200143
`
`Clinical study
`
`El-Agroudy et al, 200422
`
`Clinical study
`
`100 (kidney transplant
`patients)
`100 (kidney transplant
`patients)
`
`El-Dahshan, Bakr, Donia,
`Badr Ael, Sobh, 200625
`
`Clinical study
`
`70 (kidney transplant
`patients)
`
`Chien, Yang, Lin, Liaw,
`199744
`
`Clinical study
`
`137 (onychomycosis
`patients)
`
`El-Dahshan, Bakr, Donia,
`Badr Ael, Sobh, 200445
`Heiberg and Svejgaard,
`198146
`Soltero, Carbajal,
`Rodriguez-Montalvo,
`Valdes, 200326
`
`Clinical study
`
`Case report
`
`Clinical study
`
`70 (kidney transplant
`patients)
`1 (dermathophytosis
`patient)
`11 (kidney transplant
`patients)
`
`Ketoconazole, 21.6
`Control, 24.9
`Ketoconazole
`Baseline, 22
`6 months, 25
`12 months, 21
`10 year, 29
`Control
`Baseline, 23
`6 months, 30
`12 months, 24
`10 years, 33
`At the start of the study
`Ketoconazole, 21.31
`Control, 21.14
`
`Ketoconazole complete follow-up
`before, 13.4
`after, 19.1
`Ketoconazole incomplete follow-up
`before, 14.9
`after, 6.5
`Ketoconazole, 19.2
`Control, 20.29
`90 days, 290
`250 days, 250
`Average data, 32
`
`A recent evaluation of the FAERS database between
`2004 and 2011 (identified 1964 cases of liver injury with
`112 acute liver failure) showed that all systemic an-
`timycotics (not just ketoconazole) have been associated
`with a clinically significant risk of DILI when polyphar-
`macy and comorbidities are present.28 An analysis of
`the Kaiser Permanente Northern California database
`between 2004 and 2010 identified a similar absolute
`risk and comparable rates for ALT activity elevation
`> 200 U/L and severe acute liver injury in people receiv-
`ing fluconazole, ketoconazole, and itraconazole.29 Both
`risks are increased with voriconazole and posaconazole,
`possibly because of the more seriously ill underlying
`patient population, although the number of users of
`these drugs are relatively small.29 It was also reported
`in a review of published evidence from epidemiologic
`studies of liver injury associated with the use of keto-
`conazole, that there is no known report on severe liver
`injury in healthy volunteers who participated in drug
`interaction studies with ketoconazole.30
`In the studies analyzed in this investigation in which
`ketoconazole was used as an antifungal drug, treatment
`
`duration was relatively long (median, 276 days; approx-
`imately 9 months of administration at daily doses of
`200–400 mg; 4815 patients), but there were no clinically
`significant findings indicating a higher risk of hepatic
`injury in patients receiving ketoconazole (26 articles
`reported changes in liver function tests, and 28 reported
`no changes or did not describe this adverse effect). Four
`studies reported fatal cases because of ketoconazole.
`In the first report the patient was reported to be
`taking ketoconazole for 66 days at a dose of 200 mg
`daily coadministrated with hydrochlorothiazide, cotri-
`moxazole, and isosorbide dinitrate, and 10 days after
`discontinuation of ketoconazole the patient died. In
`this case the liver biopsy identified massive hepatic
`necrosis.3 In a second case report, 2 patients who
`received ketoconazole (200 mg/day) for 103 and 17 days,
`both died.31 In these cases no other medication was
`being taken, and the liver biopsy showed centrilobular
`necrosis in both cases.31 The third study was a literature
`review, and these authors reported 3 fatal cases that
`were considered drug related. In the first of these
`fatal cases, the patient was reported to be receiving
`
`6
`
`

`

`Outeiro et al
`
`1209
`
`ketoconazole 400 mg daily (but no details on the dura-
`tion of treatment); a biopsy showed severe hepatitis.32
`The second patient was reported to be taking keto-
`conazole 200 mg daily and also received drug treatment
`for a cardiac disease. Necrosis was not evident in
`the liver biopsy performed for this patient.32 In the
`third fatal case the person was reported to be taking
`ketoconazole 400 mg daily for 28 days; after 8 weeks
`the patient developed malaise and then jaundice. Liver
`function tests suggested liver damage, and 5 weeks
`after onset of jaundice the patient died from hepatic
`failure. There was no description of any concomitant
`medication.32 In their study Kao et al described DILI
`occurring in 28 of 57 321 patients receiving ketocona-
`zole; 1 patient who was also receiving fluconazole at the
`time died. The duration of administration was around
`90 days.33
`In 2 of these fatal cases, the role of ketoconazole as
`a contributing factor cannot be established.
`In the drug interaction studies involving 2355 partic-
`ipants (healthy volunteers and patients), 40 participants
`were reported to have increased liver transaminase ac-
`tivity (1.7%), and no deaths were reported or associated
`with ketoconazole. In the recently published analysis
`of drug interaction studies between 2005 and 2015,
`53 studies involving 971 participants exposed to sys-
`temic ketoconazole revealed evidence of ketoconazole-
`associated hepatic injury in 0.41% of participants.11
`In the studies in which patients were treated with
`ketoconazole for a longer period compared with the
`drug interaction studies, 8.7% showed elevated liver
`function tests. However, median treatment duration
`was 276 days. In contrast, the frequency of DILI in
`2 epidemiological studies with ketoconazole was low
`(2:105227 and 4.9:10 00033). The majority of patients
`received treatment with ketoconazole for less than 30
`days,27 with an average of 90 days.33 More importantly,
`the rate of DILI was low in drug interaction studies
`in which ketoconazole was administered for a short
`duration only (5 to 7 days). In the 161 studies analyzed,
`the rate of DILI was 1.7%, and there were no clinically
`significant changes in other laboratory values reported.
`There are 3 possible alternatives to ketoconazole
`discussed as a strong reference inhibitor of CYP3A4
`for drug interaction studies: clarithromycin, ritonavir,
`itraconazole.12,34,35
`Clarithromycin-mediated
`and
`CYP3A inhibition is weaker
`than ketoconazole-
`mediated inhibition; itraconazole has a long half-life,
`an active metabolite that shares the parent drug’s
`CYP3A-inhibitory properties and therefore needs
`at least 3 days of preexposure in drug interaction
`studies.34,35 According to Greenblatt, ritonavir is
`preferable to itraconazole in drug interaction studies;
`however,
`in this consideration the role of
`safety
`aspects was not clearly discussed.13,35 Regarding
`
`drug-induced liver injury, itraconazole poses the same
`risk as ketoconazole according to epidemiological
`data.28,29 Still none of
`these alternative inhibitor
`drugs can achieve the same strong inhibition as
`ketoconazole because each inhibitor has a unique
`transporter
`inhibition profile, and none of
`the
`alternative drugs provide a clean inhibition profile,
`as seen with ketoconazole.34 The majority of
`the
`alternative inhibitors are not fit for the purpose in
`exploratory drug interaction studies,34 as each of these
`alternative drugs has also been implicated in causing
`hepatic injury and other serious adverse effects.30
`For instance, itraconazole induced mild to transient
`elevation in liver enzyme activity.12 For ritonavir in
`therapeutic doses there are reports of hepatotoxicity,
`pancreatitis, and lipid disorders,36 but according
`to Greenblatt and Harmatz, in low doses (100 mg
`twice
`daily)
`ritonavir
`presents
`no
`clinically
`significant problems.13 A challenge with the use
`of clarithromycin as a prototype inhibitor is that it
`requires too many doses to achieve the necessary extent
`of inhibition.36 Furthermore, in a population-based
`nested case-control study using heath care data, the
`incidence of acute liver injury of 3.95 per 100 000
`exposures was reported for clarithromycin.37 One third
`of the cases in this study were reported to occur within
`4 days of commencing clarithromycin treatment.37
`All drugs currently considered as an alternative to
`ketoconazole in drug interaction studies carry the risk
`of acute liver injury. This risk appears to be similar
`for all these drugs including ketoconazole based on
`current information. Finally, Ke et al concluded that
`the “potential risks to DDI study subjects are low
`and manageable”36
`for
`clarithromycin-associated
`and itraconazole-associated hepatotoxicity. A logical
`conclusion of these statements is that the same low
`risk must be concluded for ketoconazole. Whether low-
`dose ritonavir poses a similar risk to healthy volunteers
`remains to be seen, but it might be a favorable option.
`A limitation of the present investigation is that
`our search strategy may not have identified all articles
`that investigated ketoconazole in drug interaction or
`antifungal treatment studies. Furthermore, the authors
`did not have full access to all the articles, and the
`abstracts do not always provide adequate information.
`Despite these limitations, the number of articles/studies
`including the present analysis is large, and we assume
`that there will be no major changes in the rate of
`DILI. Especially in the older interaction studies from
`the 1980s or 1990s it was not standard to report
`safety results in the publication. Therefore, we cannot
`be certain that liver injury was actually looked for
`in these studies because of underreporting. The rate
`of ketoconazole-associated liver injury was estimated
`to be 0.41% in the drug interaction studies between
`
`7
`
`

`

`1210
`
`The Journal of Clinical Pharmacology / Vol 56 No 10 2016
`
`2005 and 2015,11 showing the safety of ketoconazole in
`drug interaction studies conducted according to today’s
`standards. Drug interaction studies are not specially
`designed to capture elaborate safety data. However,
`at least in recent years there has also been increasing
`awareness of drug safety and toxicity in drug interac-
`tion studies, as legal authorities and ethics committees
`require more frequent safety assessments throughout
`clinical studies. For future studies it should also be a
`requirement for publication to have a safety paragraph
`included in which details about the frequency of safety
`assessments and the resulting ADRs must be presented.
`
`Conclusions
`Based on the data from more than 2000 healthy vol-
`unteers enrolled in DDI studies, the risk of hepatic in-
`jury from exposure to ketoconazole during short-term
`administration is estimated to be very low. Therefore,
`ketoconazole should continue to be used as a CYP3A
`index inhibitor in drug interaction studies in healthy
`volunteers. The use of other inhibitors will probably
`not reduce the already low frequency of elevated liver
`enzymes caused by short-term ketoconazole or might
`come with other safety issues.
`
`Acknowledgments
`We thank Prof. Andrew McLachlan for review and editing the
`English-language style.
`
`Declaration of Conflicting Interests
`None to declare.
`
`References
`1. Allen D, Wilson D, Drew R, Perfect J. Azole antifungals: 35
`years of invasive fungal infection management. Expert review of
`anti-infective therapy. 2015;13(6):787–798.
`2. Greenblatt DJ. The ketoconazole legacy. Clin Pharmacol Drug
`Dev. 2014;3(1):1–3.
`3. Lewis JH, Zimmerman HJ, Benson GD, Ishak KG. Hepatic
`injury associated with ketoconazole therapy. Analysis of 33
`cases. Gastroenterology. 1984;86(3):503–513.
`4. Hay RJ. Ketoconazole: a reappraisal. Br Med J. 1985;290
`(6464):260–261.
`5. European Commission. Ketoconazole HRA 200 mg tablets.
`2014.
`http://ec.europa.eu/health/documents/community-
`register/2014/20141119130013/anx 130013 en.pdf.
`Accessed
`June 29, 2016.
`6. FDA US. FDA Drug Safety Communication: FDA limits
`usage of Nizoral (ketoconazole) oral tablets due to poten-
`tially fatal
`liver injury and risk of drug interactions and
`adrenal gland problems. 2013. http://www.fda.gov/downloads/
`Drugs/DrugSafety/UCM362444.pdf. Accessed July 5, 2015.
`7. Comte-Perret S, Zanchi A, Gomez F. Long-term low-dose
`ketoconazole treatment in bilateral macronodular adrenal hyper-
`plasia. Endocrinol Diabetes Metab Case Rep. 2014;2014:140083.
`8. EMA. Suspension of marketing authorisations
`for oral
`ketoconazole