Clin, Pharmocokinef. 1997 May; 32 (5): 382-402 0312-5963/97/0005-0382/$ 10.50/0

DRUG DISPOSITION

© Adis International Limited. All rights reserved .

Clinical Pharmacokinetics of Tretinoin Mario B. Regazzi,1 Isabella Iacona, l Cristina Gervasutti,1 Mario Lazzarino2 and Salvatore Toma 3 1 Department of Pharmacology, IRCCS-S. Matteo Hospital and University of Pavia, Pavia, Italy 2 Institute of Haematology, IRCCS-S. Matteo Hospital and University of Pavia, Pavia, Italy 3 National Institute for Cancer Research-1ST Institute of Oncology, University of Genova, Italy

Contents Summary ................. . 1. Physicochemical Properties of Tretinoin . 2. Analytical Methods . 3. Physiological Aspects 3.1 Overview . . . . . 3.2 Intestinal Uptake 3.3 Oxidation of Retinol to Tretinoin in Tissues 3.4 Cellular Retinoic Acid-Binding Proteins 3.5 Nuclear Receptors 4. Pharmacokinetics 4.1 Absorption..... 4.2 Distribution . . . . . 4.3 Metabolism and Elimination . 4.4 Metabolism Induction . . . . 4.5 Role of CRABP in Tretinoin Pharmacokinetic Induction 5. Modulation of Tretinoin Pharmacokinetics . . . . . 5.1 Intermittent Regimen . ........... . 5.2 Administration of Cytochrome P450 Inhibitors 5.3 Liposome Formulation of Tretinoin 6. Conclusions . . . . . . . . . . . . . . . . . . .

Summary

382 384 384 385 385 386 386 386 387 388 389 391 392 393 394 395 395 396 398 399

Recent reports of the dramatic antitumour effect of tretinoin (all-trans retinoic acid) in patients with acute promyelocytic leukaemia (APL) have generated a great deal of interest in the use of this drug as a chemopreventive and therapeutic agent. However, the biological efficacy of tretinoin is greatly impaired by (presumably) an induced hypercatabolism of the drug leading to reduced tretinoin sensitivity and resistance. Several pharmacokinetic studies have shown that plasma drug exposure las measured by the plasma area under the concentration-time curve (AUC=)] declines substantially and rapidly when the drug is administered in a long term daily tretinoin regimen. These observations led to the hypothesis that the rapid development of acquired clinical resistance to tretinoin may have a pharmacological

Clinical Pharmacokinetics of Tretinoin

383

basis and result from an inability to present an effective drug concentration to the leukaemic cells during continuous treatment. The principal mechanisms proposed to explain the increased disappearance of tretinoin from plasma include: (i) decreased intestinal absorption; (ii) enhanced enzymatic catabolism; and (iii) the induction of cytoplasmic retinoic acid binding proteins (CRABP), which leads to increased drug sequestration. The most favoured explanation is that continuous tretinoin treatment acts to induce drug catabolism by cytochrome P450 (CYP) enzymes. Several strategies aimed at preventing or overcoming induced tretinoin resistance have been, and are being, planned. These strategies include intermittent dose administration, administration of pharmacological inhibitors of CYP oxidative enzymes, combination with interferon-a and intravenous administration of liposome-encapsulated tretinoin. As these strategies are now under investigation and the number of patients enrolled is small, further studies are needed to determine the efficacy and toxicity of these new schedules of drug administration. In this article we provide an overview of the relevant aspects of tretinoin physiology and pharmacokinetics, and summarise the current status of knowledge to help in the better optimisation of tretinoin administration. Tretinoin (all-trans retinoic acid) is an oxidation product in the physiological pathway of retinol metabolism which executes retinol (vitamin A)-dependent functions, including embryonic development, the maintenance of epidermal differentiation, testicular function, and regulation and differentiation of many different cell types. The existence of a cellular receptor protein for tretinoin in all vitamin A target tissues, and the evidence from several in vitro studies which indicate that tretinoin is a differentiating agent 100 to 1000 times more potent than retinol itself, suggests that tretinoin could be the active form of retinol in many physiological processes. 111 A great deal of interest has recently focused on the use of tretinoin and other retinol derivatives (retinoids) as chemopreventive and therapeutic agents. From a clinical point of view tretinoin needs to be distinguished from 13-cis-retinoic acid. The activity and effectiveness of tretinoin is well consolidated as a mono therapy regimen in the treatment of acute promyelocytic leukaemia (APL), with objective remissions ranging from 84 to 87%;1 2 -5 1 in combination with interferon-a-2a (IFNa-2a) activity has been demonstrated in mycosis fungoides and Kaposi's sarcoma.!!,·71 13-Cis-retinoic acid has been shown to be effective in the treatment of some © Adis Inte rnational Limited. All rights reserved.

preneoplastic lesions of the oral cavity, larynx and skin, and in the prevention of secondary tumours in head and neck squamous cell carcinoma. The therapeutic activity of 13-cis-retinoic acid, in association with IFNa-2a, has been shown in the cervix, skin and non-squamous-celllung carcinoma.IS-lol At present the chemopreventive activity of tretinoin is untested; data from preclinical models seem to suggest a more evident activity of 9-cis-retinoic acid, especially when used in combination with tamoxifen.1111 The use of tretinoin and its metabolites to treat premalignant lesions and to prevent chemical carcinogenesis in experimental animals has been documented for many years, and clinical studies conducted to reverse preneoplastic lesions and achieve secondary chemoprevention in cancers of the skin, head and neck, lungs, bladder and mammary gland have recently reported promising results. 1121 Tretinoin stimulates the clonal growth of normal human myeloid and erythroid precursors in vitro, inhibits proliferation and induces differentiation of blast cells from some patients with acute myelogenous leukaemia (AML). In patients with APL tretinoin demonstrated dramatic anti leukaemic activity.13.13-15 1 However, despite continuous administration, remissions obClin. Phormacokinet. 1997 May; 32 (5)

Regazzi et al.

384

tained with tretinoin alone are short-lived and patients relapse within a few months, with tretinoin failing to induce a second remission.l l6 ,l71 These results have suggested that resistance to the antileukaemic effects of tretinoin are acquired during drug therapy. Several pharmacokinetic studies have shown that plasma drug exposure, as measured by the plasma concentration-time curve (AUC=), declines substantially and rapidly when the drug is administered on a long term daily regimen.l 1X -20 ] The mechanisms proposed to explain the increased drug clearance include: • the induction of tretinoin-metabolising enzymes [particularly cytochrome P450 (CYP) enzymes] • the alteration in either the absolute amount or binding affinity of cytoplasmic retinoic acid binding proteins (CRABP) which would facilitate the transport of tretinoin to metabolising enzymes • the downregulation of the absorption process. However, since the exact pathogenesis of acquired resistance is not known, altered expression or mutation of tretinoin receptors may account for this phenomenon. 121 ,22 1 The rapid disappearance of tretinoin during long term treatment is not specific for APL. Several strategies are now under investigation for the prevention of the induction of accelerated tretinoin catabolism or to overcome it after onset. These include coadministration of inhibitors of the CYP enzyme oxidative system, different dosages of the drug, an intermittent administration regimen, the combination of tretinoin with IFNa, liposome formulation (intravenous administration of liposomal encapsulated tretinoin) and the design of new retinoids which do not bind to CRABP. These clinical pharmacological observations raise questions vital to the further exploitation of the potential of tretinoin. These questions include how to prevent the induction of the accelerated metabolism, that is considered one of the major mechanisms responsible for acquired clinical resistance to tretinoin. © Adis International Limited. All rights reserved.

~COOH Fig. 1. Structural formula of tretinoin.

This article reviews the relevant literature regarding the disposition of tretinoin and the strategies to modulate its pharmacokinetics.

1. Physicochemical Properties of Tretinoin The structure of tretinoin, as with other naturally occurring retinoids, can be divided into 3 moieties: a polar terminal end (carboxylic acid), a conjugated side-chain and a trimethylcyclohexenyl terminal group (fig. I). Tretinoin occurs as yellow to light orange crystals or crystalline powder with the odour of silage (ensilage or hay). It should be stored in tight, lightresistant containers at 15 to 30°C; it melts at between 176 and 181°C. The molecular weight is 300.42. Tretinoin is insoluble in water and slightly soluble in alcohol and chloroform. Tretinoin is prepared by the oxidation of a retinal aldehyde group (obtained by oxidising retinol) to a carboxyl group. Tretinoin is an antioxidant and free radical scavenger.

2. Analytical Methods Tretinoin is rapidly isomerised and destroyed in the presence of light and oxidants, therefore considerable attention must be given to the collection, handling and storage of biological samples. It is particularly important that daylight be excluded from the laboratory while tretinoin and other retinoids are being analysed. Gold fluorescent lighting of moderate intensity does not cause any isomerisation and such lighting is recommended for retinoid laboratories. Stock tretinoin solutions should be stored in the dark at low temperature in an oxygen-free atmosphere and should be periodically checked for deterioration. Investigation into the pharmacokinetics and metabo
ism of tretinoin has encountered many diffiClin. Pharmacokinet. 1997 May; 32 (5)

Clinical Pharmacokinetics of Tretinoin

cuI ties because of the lack of suitable techniques for dealing with this labile compound. The thermal instability of tretinoin excludes gas chromatographic methods; high pressure liquid chromatography (HPLC) is the most frequently used method of analysis for tretinoin. HPLC analysis requires the addition of acid and/or a salt such as ammonium acetate (ammonium formate and ammonium bicarbonate have also been used) to the mobile phase to suppress ionisation.[23· 24 1 Plasma is the most frequently analysed sample for tretinoin. Generally it is necessary to denature retinoid-binding proteins with organic solvent, usually by precipitation with methanol or alcohol (ethanol) before extraction. In most cases repeated extraction with an organic solvent is necessary, thereby resulting in a large volume of extract. Such large volumes cannot be injected into an HPLC system, and must, therefore, be concentrated or evaporated to dryness and reconstituted in an appropriate solvent before HPLC analysis. Because degradation of retinoids (especially when the concentration is very low, as in tissue extracts) is accelerated by heat, strong heating should be avoided in removing solvents from extracts. The detection of retinoids in HPLC is usually done by measuring the absorption of ultraviolet light. Because of its conjugated polyene chains, tretinoin absorbs light strongly at wavelengths of 345nm, a wavelength where few other compounds absorb light appreciably. In a very different approach, tretinoin and other retinoids are analysed from plasma without solvent extraction; instead, column switching is used to extract and concentrate these compounds from human plasma. 1251 Because of the instability of treti no in and its ease of isomerisation, in both methods the reported reference compounds and control runs must be used to obtain accurate analysis.

3. Physiological Aspects 3.1 Overview

Tretinoin is normally found at very low concentrations (approximately 4 to 14 nmol/L) within the © Adis International Limited. All rights reserved.

385

human circulation and bound to albumin. Noyl261 reported that bovine serum albumin presents 3 different sites for retinoic acid binding: 2 could be correlated to known binding sites for long chain fatty acid and I appeared to be a unique site for retinoic acid binding. It is not known whether tretinoin present in the plasma is delivered exclusively from dietary sources (from the intestine) or whether some of the circulating tretinoin is synthesised and secreted by other organs in the body. No membrane cell surface receptor specific for retinoic acid is known and at physiological pH the protonated (uncharged) form was found to be stabilised by incorporation in the bilayers and can cross the membrane rapidly and spontaneously,127 1 like other hydrophobic carboxylic acids (fatty acids and bile acids). However, any contribution of plasma component to normal physiology seems unlikely because, under homeostatic conditions, cells probably derive whatever retinoic acid they require by intracellular conversion from retinol. At the cellular level 2 distinct cell (CRABP) have been discovered and well characterised: types I and II, with CRABP I having a more widespread distribution than CRABP II. While originally thought to perform a largely sequestratory function, CRABP binding may facilitate the presentation of tretinoin to oxidative enzymes which catalyse conversion to inactive metabolites, as suggested by experiments showing that overexpression of CRABP results in a diminished response to tretinoin by murine teratocarcinoma stem cells.l28.291 Only unbound retinoic acid, which diffuses through the nuclear membrane, is potentially effective in transducing a response. Tretinoin effects seem to be mediated predominantly via binding to a group of related nuclear receptors [retinoic acid receptors (RAR-a, ~, y) and retinoid X receptors (RXR-a)] which interact with specific DNA sequences (retinoic acid response elements), and lead to the transcriptional activation of genes with appropriate retinoic acid response elements. While 9-cis-retinoic acid has a high affinity for RXR and binds efficiently to RAR, tretinoin is a weak ligand Un. Pharmacokinet. 1997 May: 32 (5)

386

for RXR receptors interacting 40 times less efficiently with RXR compared with RAR.l 301 3.2 Intestinal Uptake

Retinoids in the body originate in the diet, either as provitamin A (proretinoi) carotenoids or as preformed retinoids. The dietary carotenoids and preformed retinoids undergo a series of metabolic conversions, extracellularly in the lumen of the intestine and intracellularly in the intestinal mucosa, which results in a preponderance of the absorbed dietary retinoids being converted to retinol.J311 Only a very small proportion of dietary retinoids are converted to retinoic acid (or comes in the diet as such) and enter the circulation through the portal system, bound to serum albumin. 132 ,331 Some of the tretinoin formed in the intestine from dietary sources is excreted through the bile, although a large percentage of the tretinoin of dietary origin appears to be removed from circulation by the tissue.13 4 1Homeostatic mechanisms, currently undefined, act to control plasma tretinoin concentrations. Eckhoff et al. 1351 demonstrated that tretinoin concentrations in humans increase transiently in response to the administration of retinol, but then return to the initial steady-state concentrations with continued administration. Other dietary factors have also been reported to influence circulating levels of retinoic acid. For example, rabbits fed ~-carotene have a 2-fold higher serum concentration of tretinoin than those receiving no ~-carotene.1361 In fact the intestinal mucosa cleaves ~-carotene to retinaldehyde, which is oxidised to tretinoin and then released into the portal circulation. Napoli and Race l371 suggested that the metabolism of dietary ~-carotene is not simply a process for producing retinaldehyde, which is subsequently metabolised to either retinol or tretinoin. These authors l371 demonstrated in the intestinal mucosa, in addition to the central ~-carotene cleavage which generates retinaldehyde, an excentric cleavage to form ~-apocarotenals. Apocarotenals may be directly converted to tretinoin. Ii) Adis International Limited. All rights reserved.

Regazzi et al.

3.3 Oxidation of Retinol to Tretinoin in Tissues

Most cells and tissue seems to have the metabolic machinery needed to oxidise retinol to retinoic acid, but the biochemical process by which retinoic acid is enzymatically formed within tissues has not been unequivocally established. The currently prevailing hypothesis is that retinol is first oxidised to retinaldehyde, which in turn is oxidised to retinoic acid. This process is thought to be similar to the oxidation of alcohol to acetaldehyde, which is in turn oxidised to acetic acid. Therefore, the same enzymatic system (alcohol dehydrogenase-aldehyde dehydrogenase-aldehyde oxidase) responsible for acetic acid formation could be involved in retinoic acid formation. However, several published studies do not clearly define the role of relatively nonspecific enzymes such as alcohol dehydrogenase, aldehyde dehydrogenase or aldehyde oxidase in the conversion of retinol to retinoic acid. The majority of recent data[881 would suggest that the oxidation of retinol to retinaldehyde is probably catalysed by a microsomal enzyme or enzymes which use retinol bound to specific cellular retinolbinding proteins (CRBP) as their substrate. Whether the subsequent oxidation to retinoic acid is also CRBP-dependent, or is catalysed by a soluble aldehyde dehydrogenase or aldehyde oxidase or both, remains uncertain.[39[ In spite of the equivocal data available, multiple enzymatic activities are clearly involved in the conversion of retinol to retinoic acid. 3.4 Cellular Retinoic Acid-Binding Proteins

Because of the hydrophobic nature and lability of retinoids, specific carrier proteins, which are involved in their transport and metabolism, are necessary. To date, 2 distinct intracellular retinoic acid binding proteins have been discovered and well characterised: CRABP I and II. CRABP I and II specifically bind to retinoic acid but not to retinol or retinaldehyde. The determination of CRABP distribution in organs of various species is diffiClin. Pharmacokinet. 1997 May: 32 (5)

Clinical Pharmacokinetics of Tretinoin

cult, as intraspecies differences very probably exist in the levels of CRABP for a given organ. In rats CRABP I was found to be particularly abundant in the male and female reproductive tract including the testis, epididymis, vas deferens and seminal vesicles, and in ovary, uterus and oviduct. 1401 Relatively high concentrations were also found in the brain and spinal cord. In other studies the retina and pigment epithelium of the bovine eye l411 and the bovine adrenal gland l421 were found to be rich in CRABP I. However, we know still little about the distribution of CRABP in human organs although the uterus apparently has a high concentration. 1431 The presence and level of CRABP I in the skin (an organ particularly responsive to tretinoin) appears to vary among different species. Adult human epidermis did not have detectable levels of mRNA for CRABP I, but only that for CRABP 11.1441 It is presently thought that CRABP II is expressed most strongly during embryogenesis, when it is more widely distributed. No species reliant on retinol has been found to be lacking such proteins. CRABP protein may function in tretinoin storage, transport to the nucleus or to the endoplasmic reticulum (where it is metabolised), thus modulating the intracellular concentrations. In fact, CRABP are probably not simply passive reservoirs but may function as buffers which maintain low, tolerable concentrations of free tretinoin, the metabolically or functionally important pool. A passive role would not appear to require the existence of 2 intracellular binding proteins and does not explain the very high degree of identity of amino acid sequence retained for the homologous proteins of the various species examined. This conservation suggests that the surface of the protein, as well as its tretinoin-binding site, is important to its functions, and implies that the recognition of the protein by other cellular components is probably part of the process of overall retinoid trafficking. 3.5 Nuclear Receptors The physiological effects of tretinoin are mediated through binding to nuclear receptors, which reV Adis International Limited. All rights reserved.

387

suit in a modulation of the expression of specific gene products. To date, 2 families of nuclear retinoic acid receptors have been described. In the late 1980s, 3 closely related tretinoin receptors (RARa, ~ and y) were identified as members of the superfamily of nuclear receptors.f45I Structurally these receptors are characterised by 2 major functional domains, one of which binds DNA and the other hormone. The domain which binds hormone also contains a region involved in receptor dimerisation, a function which markedly increases DNAbinding affinity. Activated RARs bind to specific sites in DNA and act as transcription factors, ultimately controlling the expression of target genes. The ligand-binding domains of the RARs are highly conserved (>75% amino acid identity), suggesting that they arose from a common ancestral RAR gene. Within each receptor subtype ofthe RAR family there are splice variants which give rise to multiple isoforms. Seven isoforms of RAR-a and multiple forms of RAR-~ and RAR-y have been described. RAR isoforms are expressed in distinct patterns throughout development and in the mature organism, indicating that they may mediate different functions. In addition, 3 other receptors of the 'X' family (RXR-a, ~ and y) have been identified. 1461 RXRs are only distantly related to the RARs in the protein sequence. Tretinoin and its isomer 9-cis-retinoic acid can directly bind the RAR family with similar affinity. It can also directly bind and activate RXRa; in contrast, tretinoin interacts 40-fold less efficiently with RXR-a than with RAR-a. The pleiotropic effects of retinoids must be partly explained by the patterns of expression of various RAR and RXR genes. 147 ,4X 1 APL is characterised at the molecular level by a t( 15; 17) chromosomal translocation, which fuses the promyelocytic myeloid leukaemia (PML) gene to the RAR-a gene. It is believed that the PMLRAR-a fusion protein plays a crucial role in APL pathogenesis by blocking the promyelocyte in early proliferative stage of its differentiation. The PML-RAR fusion protein is present to a large exClin. Pharmacokinet. 1997 May; 32 (5)

388

tent in leukaemia cells and, therefore, is able to exert a dominant negative effect on RAR, RXR and PML function in both the presence and absence of retinoid. Tretinoin therapy may reverse the block in promyelocytic differentiation, leading to disease remission by overcoming the dominant roles of PML-RAR-a on PML distribution and on RXR localisation. As a result, RXR heterodimers (PMLRARlRXR, RARlRXR and possibly also VDRlRXR) would be able to effect ligand-mediated transcription from key DNA target sequences involved in differentiation. Remission requires relatively high concentrations of tretinoin.[ 14,21 [ The relationship between intracellular tretinoin concentration and its efficacy is supported by many reports)3·21,39[ It may be expected that variations in tretinoin concentrations may result in reduced biological response.

4. Pharmacokinetics Detailed analysis of the pharmacokinetic behaviour of tretinoin in humans has been restricted by both the low plasma concentrations reached after an oral dose and the absence of an intravenous form which would avoid the variability associated with oral drug administration. Consequently, most pharmacokinetic data are derived from studies following oral administration, and the estimates of apparent volume of distribution (Vd) and clearance are all corrected for the unknown fraction of the drug systemically available [or bioavailability (F)]. Tretinoin pharmacokinetic parameters after single and multiple doses in humans are summarised in tables I and II, respectively. Preclinical studies using an intravenous formulation of tretinoin in a rhesus monkey pharmacokinetic model showed that tretinoin is cleared through a capacity-limited (saturable) process.[20[ The plasma area under the concentration-time curve (AUC) following an intravenous bolus dose of tretinoin presented 3 distinct phases: an initial rapid distributive phase, a relative plateau phase (the duration of which was proportional to the dose) and a rapid exponential elimination phase. The curve is most consistent with a capacity-limited saturable © Adis International Limited. All rights reseNed.

Regazzi et al.

process: at high drug concentrations the elimination process is saturated and the rate of elimination is independent of the drug (substrate) concentration. However, because of the multiple pathways of tretinoin metabolism, the profile of tretinoin elimination may derive from 2 parallel processes, a capacity-limited process and first-order elimination. Other studies had demonstrated that tretinoin pharmacokinetics were dose-dependent following a single intravenous injection in rats[ 59 1and rhesus monkeys.[60 1 EI Mansouri et al. 1591 found that the percentage of the tretinoin dose excreted in urine decreased as the dose administered increased; while at high doses the glucuronoconjugates represented a higher percentage of the excreted dose, whereas the polar metabolites were less prevalent. Animal studiesI34,59,601 documented the timedependence of tretinoin pharmacokinetics, because the concentrations and pharmacokinetic parameters varied upon repetitive administration. Induction of the capacity-limited elimination process is probably responsible for the decrease in plasma drug concentrations seen as early as the third day of daily intravenous administration. In humans, the pharmacokinetic behaviour of orally administered tretinoin is characterised by rapid drug elimination, with a terminal half-life (t l/ 2Z ) of approximately 45 minutes.[49, 50 I After oral administration of tretinoin 45 mg/m2/day to humans, the peak plasma concentrations (observed I to 3 hours later) approached the Michaelis-Menten constant (Km) calculated in the monkey and, on this basis, it was postulated that the elimination of tretinoin in patients is also dose-dependent and capacity-limited. 1611 Long term tretinoin administration appears likely to lower peak concentrations and total systemic exposure (AUC=) in studies in mice,1 59 1monkeysl201 and human patients. I19,501 The progressive reduction of tretinoin plasma concentrations with long term administration, whatever the aetiology, may be an important mechanism of acquired resistance to tretinoin therapy in patients with APL. Clin. Pharmacokinet. 1997 May: 32 (5)

Clinical Pharmacokinetics of Tretinoin

389

Table I. Mean pharmacokinetic parameters (± SD or range where available) of tretinoin in humans after a single dose Reference

AUC_ (Ilg/L· h)

CUF

Diagnosis

Lefebvre et a1. 118]

15

APL

45

590±700

1.9

630

71.4

39 ± 19

Muindi et a1. 149]

13

APL

45

346 ± 266

2.2 ± 1.0

682 ± 500

65.9 ±48.3

48 ±6

499 ±200

90.1 ±36.1

10 Muindi et al.[SO] Smith et al.[51]

45

279 ± 243

2.2± 1.0

603 ±442

74.6 ± 54.7

48± 12

Solid tumours

45

191 ± 137

3.3 ± 1.4

454±419

99.1 ±91.4

48± 18

2

Paediatric tumours

22.5

1.7'

387 ± 84

58.1 ± 12.6

58 (52-66)

8

Paediatric tumours

30

2.0'

339 ± 141

88.5 ± 36.8

45 (36-64)

8

Paediatric tumours

40

3.0'

40 (27-77)

Solid tumours

60 1038 ± 541

2365 ± 163

1396

3570

56.0

1630

733

34.1

Paediatric tumours

40

8

74.0±5.1

175

8

13

15.8

200 25

7

39.8 ± 24.9

Solid tumours APL

Lee et al. [55]

1005 ± 630 3800

Solid tumours

5 3

a

t1;"Z (min)

APL

Smith et al.[54]

Regazzi (unpublished oservations)

45

(Uh/m2)

20

2

Adamson et al.[56]

Cmax (Ilg/L)

47

Lee et al. [52]

Castaigne et al.l 53]

Dose (mg/m2)

tmax (h)

No. of patients

3.0'

17 45

Kaposi's sarcoma

40

330 ± 60

CML

40

219± 131

3

MDS

22.5

APL

45

39.8 ± 24.9

168

Solid tumours

8

1005 ± 630 1355 ± 1479 (median 1147)

2.6 ± 0.9

40 (27-77)

101.2 33.2 ± 36.2

725± 130

55.2 ± 9.9

568 ± 440

70.4 ± 54.5

66 ±2

75±42

3.3 ±2.3

190±84

118.4±52.1

54±4

321 ±204

2.6 ± 0.7

673 ± 313

66.8 ± 31.1

50± 10

Median.

= acute promyelocytic leukaemia; AUC_ = area under the concentration-time curve; Cm,x = maximum plasma concentration; CUF = total body clearance divided by bioavailability; CML = chronic myelogenous leukaemia; MDS = myelodysplasticsyndrome; tmax = time to reach maximum drug concentration; t'l,z = elimination half-life associated with the terminal slope Az of the semologarithmic concentration-time curve. Abbreviations: APL

4.1 Absorption

Tretinoin is formulated as a 10mg capsule which contains a suspension of the drug in oil: the solubility of tretinoin is limited in aqueous solutions. Absorption is dependent upon the release of the drug from the capsule and is also likely to be a function of the pH and fatty acid composition of intralumenal bile. This highlights the importance of the effect of a high-lipid diet on the absorption of tretinoin. As the oral absorption of most retinoids is increased with food, differing dietary regimens may contribute to the intra- and interpatient variation in peak plasma tretinoin concentrations reported in some studies. I ]4,391 Tretinoin bioavailability has been estimated at about 50% on the basis of studies conducted in volunteers. 1621 © Adis International Limited. All rights reserved.

Several studies in experimental animals examined the appearance of tretinoin in the circulation after intraduodenal, oral and intraperitoneal administration of radiolabelled retinoic acid.163-66J Orally administered tretinoin is mainly absorbed from the intestine into the portal route and about two-thirds of the absorbed dose becomes distributed as tretinoin throughout the body.133 J A study by EI Mansouri et al. 1591 reported that the absorption pharmacokinetics of tretinoin in rats are not simply first-order but were sometimes zero-order. The 2 standard explanations for this phenomenon are the limited solubility of the drug in the intestinal fluid and carrier-mediated transport. Since tretinoin was administered as a solution, active transport is the more likely hypothesis, which implies the existence of an intestinal carrier. Clin. Pharmacokinet. 1997 May: 32 (5)

Regazzi et al.

390

Table II. Change in the pharmacokinetic parameters of tretinoin in humans after multiple doses Reference

Smith et al.[51J

No. of patients

7

Diagnosis

Paediatric tumours

Treatment regimen

Continuous therapy

Daily

Pharmacokinetic study

dosage (mg/m2)

dose

Cmax

AUC~

(mg/m2)

(~g/L)

(~g/L' h)

45-60-90

22.5-30-45 447 ± 237

day 1

<90 ± 150

day 28 Muindi et al. [49J

7

APL

Continuous therapy

45

day 1

294±89

537 ± 191

138 ± 139

248 ± 135

Lee et al. [52J

Solid tumours

60

60

2

175

175

200

200

45

45

after2.1mo

Adamson et al. [56J

1038±541 155 ± 18

after 2.1 ± 0.1mo day 1

at relapse

742 293

after 10.5mo

day 1

APL

1396

244 ± 145 Kaposi's sarcoma

7 days onl 7 days off

40

day 15 (day 1 wk 3) day 71 (day 1 wk 11) Solid tumours 4

Adamson et al.[57J

7

14 days onl 7 days off

90

J.,61

J.,51

725± 130 90±20

J.,88

885 ± 195

122

640 ± 130

J.,12

45 1355 ± 1479 308 ± 476

Refractory cancer

3 days onl 4 days off

90

330±90

day 3

60±30

day 29 (day 1 wk 4)

300± 120 Kaposi's sarcoma

3 days onl 4 days off

45

J.,77

30

day 1

Toma et al.[58J

J.,66

40 330 ±60

6

800 ± 240

152

day 7

day 14

J.,58

499 ± 200

day 1

day 1

1590 2365 ± 162 3570

6

Lee et al. [55J

3800

890

10 8

J.,54

Continuous therapy

day 1

Muindi et al.[50J

J.,80

45

after2-6wk

day 1

reduction or increase vs day 1 (%)

AUC~

500 ±75a 130 ± 120

J.,74

570 ± 95

114

22.5

day 1

55.7

91

day 3

9.6

20

J.,78

53.6

115

126

day 22 (day 1 wk 4)

27.0

75

J.,18

day 52 (day 3 wk 8)

52.3

62

J.,32

day 8 (day 1 wk 2)

Regazzi (unpublished observations)

13

CML

7 days onl 7 days off

80

40

day 1

219 ± 131

day 7

71 ±58

192± 169

J.,66

203 ± 145

545 ± 395

14

day 15 (day 1 wk 3)

rtJ Adis International Limited. All rights reserved.

567 ±440

Clin. Pharmacokinet. 1997 May; 32 (5)

Clinical Pharmacokinetics of Tretinoin

391

Table II. Contd Reference

No. of patients

6

a

Diagnosis

APL

Treatment regimen

Daily dosage (mg/m2)

2wk on! 10wkofi

45

Pharmacokinetic study dose (mg/m2)

Cmax (!lg/L)

AUC~

(!lg/L· h)

45

day 1

321 ± 204

673 ± 313

day 14

131 ±94

245 ±201

Standard error.

=

AUC~ reduction or increase vs day 1 (%)

=

j,64

Abbreviations and symbols: APL acute promyelOCy1ic leukaemia; %AUC~ change in area under the concentration-time curve; C max maximum drug plasma concentration; CML = chronic myelogenous leukaemia; wk = week(s); j, indicates decrease; i indicates increase.

This hypothesis is consistent with the decreased F with the higher dose, a phenomenon which would be explained by the saturation of the carrier. The results of a study conducted in 5 patients with APU 531 showed a dose of tretinoin 25 mg/m2/day gave the same plasma concentrations as a dose of 45 mg/m2/day, this seems to confirm the hypothesis. However, the results of a study in 6 patients with solid tumours by Adamson et al. 1671 specifically designed to examine the saturability of tretinoin absorption, suggested that the gastrointestinal absorption of tretinoin is not saturable but rather is highly variable. But given the small number of patients studied, these results will require confirmation. Downregulation of absorption and an increased first-pass effect have been reported unambiguously in rats l591 and explain, in part, the reduction in plasma tretinoin concentration during long term administration. However, the lack of available data on both oral and intravenous routes of administration and the unknown extent of intrapatient variability in the absorption of tretinoin precluded a similar conclusion in humans.

4.2 Distribution After oral administration, tretinoin is rapidly and extensively distributed to tissues, but was not detected in the cerebrospinal fluid and thus is not likely to be an effective treatment for leukaemic symptoms in the eNS. As with physiological tretinoin, it appeared to be extensively bound. (j)

Adis International Limited. All rights reserved.

=

Tretinoin normally circulates bound to serum albumin.16g1 The lack of a specific transport protein may explain why the concentrations of tretinoin vary over a 2-fold range or more during a single day. Tretinoin is not stored in any tissue, as confirmed by experimental studies, indicating that the total recovery of radioactive compounds in the entire body represents low percentages of the total administered label. Between the tissues examined, the liver, kidneys and intestine have relatively high concentrations, and serum and adrenal glands have intermediate values, whereas the testes and fat-pads have the lowest tretinoin concentrations.133,691 Rabbit bone is known to be an important metabolic and storage site for tretinoin,1 70 I which appears to be essential for the normal differentiation of leucocytes.14.7 J ,721 This observation may be an important physiological link between the delivery of retinoids by chylomicrons to bone marrow and the likely functional role played by retinoids in the regulation of blood cell differentiation . The inability to store tretinoin and its analogues may be an advantage when it is used as therapy. In fact, liver toxicity, which occurs with high concentrations of retinol and retinyl esters, is infrequently seen with tretinoin treatment. Sandberg et al. 1731 studied the intravenous pharmacokinetics of tretinoin in the cynomolgus monkey. In this study the apparent V d of tretinoin was dose-dependent and significantly greater at low doses (V d 0.52 ± 0.12 Llkg after 0.0125 mg/kg, Clin. Pharmacokinet. 1997 May: 32 (5)

392

0.21 ± 0.05 Llkg after 0.25 mg/kg; p < 0.05). The time-concentration profile was fitted by a 2-compartment open model and suggested that less than 30% of tretinoin was available in the central compartment for elimination in the postdistribution phase. In the monkey, tretinoin showed a slow equilibration and binding in the peripheral compartment of the 2-compartment model. In humans there is limited pharmacokinetic information on the Vd of tretinoin as the most unambiguous means of determining this parameter is by intravenous administration. Because only the oral pharmacokinetics of treti no in have been examined, most information has been limited to AUC= and

4.3 Metabolism and Elimination

Tretinoin is rapidly metabolised in vivo and in vitro to a variety of oxidised and/or conjugated metabolites. Metabolites are excreted as glucuronide conjugates in the urine and bile. About 30% of the total radioactivity was excreted in the urine and 60% in the faeces after a single dose of radiolabelled tretinoin.1691 Metabolites of treti no in generated in vivo include: stereoisomerisation derivatives (9-cis-retinoic acid and 13-cis-retinoic acid), oxidation derivatives (4-hydroxy-retinoic acid, 4-oxo-retinoic acid, 18hydroxy-retinoic acid, 5,6-epoxy-retinoic acid, 3,4-didehydro-retinoic acid and retinotaurine), stereoisomerisation and oxidation derivatives (13cis-4-oxo-retinoic acid), glucuronidation derivatives (retinoyl ~-glucuronide, 13-cis-retinoyl ~-glucu­ ronide, 4-oxo-retinoyl ~-glucuronide, 5,6-epoxyretinoyl ~-glucuronide and 13-cis-4-oxo-retinoyl ~-glucuronide) , nonpolar metabolites of retinoic acid and retinoic acid esters. Some of these metabolites are active in mediating tretinoin function, whereas others are probably catabolic products. 1741 Experimental studies indicate that the major pathway of tretinoin metabolism consists of the hydroxylation at position C-4 of its cyclohexenyl ring following the sequence retinoic acid ~ 4-hydroxy-retinoic acid ~ 4-oxoretinoic acid ~ more polar metabolites.175,761The © Adis International Limite d . All rights reserved.

Regazzi et al.

first and last step of this sequence require NADPH, whereas the oxidation of 4-hydroxy- to 4-oxoretinoic acid is NAD+ (or NADP+) dependent. Both NADPH-dependent steps, but not the NAD+-dependent dehydrogenase reaction, require oxygen and are strongly inhibited by carbon monoxide and other CYP inhibitors such as Iiarozole and ketoconazole.[77-80 1 The enzymatic activity required for the formation of 4-hydroxy-retinoic acid and the metabolism of 4-oxo-retinoic acid to more polar metabolites have been attributed to CYP in the microsomal fraction obtained from various tissues. It is surprising that different CYP isozymes are able to catalyse the 4-hydroxylation of retinoic acid. Many liver CYP isoforms, including 2A4, IA2, 2EI, 2E2, 2C3, 2GI (and 3A6 in rabbits)181 1 and 2B I (and 2C7 in rats) can catalyse the 4hydroxylation of retinoic acid. The isozyme CYP2C8 of human liver microsomes was shown to metabolise retinoic acid, in a reconstituted system, to more polar metabolites, including the corresponding 4hydroxy compounds.1821 It is probable that lipoxygenase, an enzyme which oxidises polyunsaturated fatty acids, also participates in oxidative tretinoin metabolism and, therefore, in the 4-hydroxy-retinoic acid and 4-oxoretinoic acid formation . A relationship between CRABP, retinoic acid metabolism and retinoic acid potency has emerged. The metabolism of retinoic acid, bound to CRABP, is about 7-fold more efficient than that of free retinoic acid. CRABP seem to attenuate retinoic acid action by sequestering retinoic acid and acting as a conduit for its efficient metabolism. As predicted by their affinities , several metabolites were bound to CRABP during tretinoin metabolism. Tretinoin was metabolised readily in the presence of 2 equivalents of CRABP, but similarly bound 4-hydroxy-retinoic acid and 4-oxo-retinoic acid were shielded from rapid metabolism. 1561 All-trans derivatives of tretinoin (3,4-didehydroretinoic acid, 4-hydroxy-retinoic acid, 4-oxoretinoic acid and 18-hydroxy-retinoic acid) have affinities similar to that of tretinoin in binding to Clin . Pharmacokinet. 1997 May; 32 (5)

Clinical Pharmacokinetics of Tretinoin

CRABP, whereas cis-derivatives (l3-cis-retinoic acid and 13-cis-4-oxo-retinoic acid) have affinities an order of magnitude lower. These data suggest a mechanism for discriminating between the metabolism of all-trans and cis-retinoids.f741

4.4 Metabolism Induction The decrease in plasma tretinoin exposure over time has been observed in several pharmacokinetic studies and was initially noted in patients with APL,f18.19,49,501 Pharmacokinetic studies of APL patients who relapsed while on therapy with tretinoin 45 mg/m2/day revealed a marked decrease (",,50%) in plasma AUC~ compared with day I values (499 ± 200 Ilg/L • h on day I vs 244 ± 145 Ilg/L. h at the time of relapse). Doubling the dose of treti no in in these patients not only failed to increase AUC~ (222 ± 112 Ilg/L. h, P = 0.56), but also failed to induce a second remission. 1191 1t was concluded that the eventual occurrence of resistance to tretinoin therapy is caused by progressively decreasing plasma drug concentrations. There is now evidence that APL cells at relapse show reduced sensitivity to low tretinoin concentrations in vitro, and it may also be suggested that failure to respond to tretinoin may be due to a pharmacological inability to present effective drug concentration to the leukaemic cells.1831 Increased oxidation via the CYP enzyme system (lO-fold increase in urinary 4-oxo metabolites) appears to account, at least in part, for the observed reduction in plasma drug concentrations and development of resistance. The autoinduction phenomenon, which was first described by Roberts et aLI 75 1in hamster intestine and liver, was shown to occur both in small intestinal mucosa and in liver, but not in epithelial tissues. Roberts et aL175, 78 1 observed that tretinoin was rapidly metabolised by liver microsomes, either from retinol-normal hamsters or from retinoldeficient hamsters, which had been pretreated with tretinoin , but not by microsomes from retinoldeficient animals. In direct contrast, the rate of metabolism of 4-hydroxy-retinoic acid was equivalent in each of these microsomal preparations, It might © Adis International Limited. All rights reserved.

393

be considered that the induction of tretinoin catabolism represents a specific adaptation to increase its rate of elimination. It is easy to deduce that not only does tretinoin pretreatment affect metabolic velocity, but also retinol status is a determining factor in the rate of excretion of polar metabolites when physiological doses of tretinoin are used. Skare and Deluca l63J demonstrated that when physiological doses of tretinoin are administered to rats, a delay of 4 to 6 hours was observed before the appearance of metabolites in the bile of retinoldeficient rats compared with normal rats, This delay may result from the loss of retinoid-induced enzymes normally used to metabolise tretinoin to more polar metabolites. An alternative explanation for the delay in excretion could be that tretinoin is transported to target organs in need of retinol, or saturates retinol-binding proteins. After pretreatment of neonatal rats with tretinoin, the rate of 4-hydroxylation increased about 4 times,l 84 1 Other drugs which may induce CYP, such as phenobarbital, pregnenolone J6acarbonitrile or troleandomycin (triacetyloleandomycin) pretreatment, increased tretinoin metabolism to a much lesser degree.f 851 To better understand the basis for the decrease in plasma drug concentrations, Adamson et aL120,601 performed a series of pharmacokinetic studies using an intravenous formulation of tretinoin in rhesus monkeys. The pharmacokinetics of tretinoin were dose-dependent and drug elimination was consistent with a capacity-limited process. Induction of this capacity-limited elimination process was responsible for the decrease in plasma drug concentrations observed with long term daily drug administration. A decrease in plasma concentrations was noted as early as the third day of daily intravenous drug administration, Administration of CYP inhibitors such as ketoconazole and Iiarozole can increase the plasma concentrations of tretinoin in animal models and in humans,15(),1l6, 87 1 suggesting that oxidation of tretinoin is one pathway for its catabolism, despite the fact that different studies have tentatively idenClin. Pharmacokinet. 1997 May; 32 (5)

394

tified the 4-oxo-all-trans-retinoic acid glucuronide as the only urinary metabolite increased with long term tretinoin administration. Because both oxidation or glucuronidation can be upregulated, the capacity-limited process induced with long term drug administration may be either oxidation or glucuronidation. Many drugs undergo oxidative followed conjugation metabolic reactions, and if oxidation of tretinoin is the rate-limiting step in its metabolism, significant plasma/urine concentrations of 4-oxo-all-trans-retinoic acid would not be observed even following enzyme induction because of the subsequent, and presumably more efficient, glucuronidation step. Moreover, in patients treated with tretinoin the decrease in plasma drug exposure observed occurs not over weeks or months, but within days of the initiation of drug administration. By the end of the first week, the plasma AUC= decreases to approximately 15% of baseline. The time course of changes in plasma tretinoin concentrations observed in the rhesus monkey following long term administration of the drug appears to parallel that observed in humans, so it is reasonable to hypothesise that the decrease in plasma tretinoin concentration in patients is also the result of induction of the metabolic pathway. The progressive decrease in plasma AUC= values with continued oral therapy is not APLspecific, as it was observed in patients with other diseases.160.52.55.57.58,881

4,5 Role of CRABP in Tretinoin Pharmacokinetic Induction

Another factor which might contribute to the retinoid relapse phenomenon involves the role of the high affinity retinoic acid-binding proteins CRABP I and II. These proteins are believed to mediate the transfer of the retinoid form from cytoplasm to the nucleus of the celL Increased levels of CRABP may cause the pooling of retinoids in tissues, resulting in low plasma concentrations and accelerated clearance of the drug from the circulation,174,891 In normal body tissues the expression of © Adis International Limited. All rights reserved.

Regazzi et al.

CRABP is thought to increase with continuous exposure to retinoids,1 90 I An overall increase of tretinoin catabolism in cell tissues by an increase of CRABP, as has been shown in the skin and haemapoietic tissue,[28,29, 91 1 may account for the low plasma tretinoin concentrations observed during continuous tretinoin therapy and for the failure to achieve effective tretinoin intracellular concentrations, This altered in situ cellular metabolism of tretinoin and induced hypercatabolic state leading to lower plasma concentrations of tretinoin may combine to cause acquired tretinoin resistance. Since the liver appears to be the primary site of tretinoin biotransformation, and because it expresses relatively little CRABP, it is not clear what role is played by CRABP in the hepatic metabolism of the drug. Indeed, the increase in plasma drug clearance noted with continuous tretinoin administration appears to be related to metabolic enzyme induction and not to the increase of CRABP expression, as suggested by the differences in the time course of both the up and downregulation of tretinoin clearance compared with CRABP skin levels. 1201 The amount of total CRABP (I and II) measured in skin biopsy samples and the values of tretinoin plasma clearance, plotted as a function of days of drug administration, are shown in figure 2. In their study with rhesus monkeys, Adamson et aL[ 20 1 showed that the amount of CRABP measured in skin biopsies increased rapidly during continuous tretinoin administration by intravenous bolus, reached a peak of approximately 3 times the baseline concentrations by day 3 of tretinoin administration and remained at this level for the remainder of the period of long term drug administration. After 7 days without therapy, CRABP levels had decreased but had not returned to baseline levels, Therefore, it has been suggested that 2 parallel, but potentially related, events may reduce the utility of tretinoin when administered long term: • an increase of drug plasma clearance followed by a decrease in drug plasma concentration Clin. Pharmacokinet. 1997 May: 32 (5)

Clinical Pharmacokinetics of Tretinoin

395

o

c

•

~ 1000

~~

Ci 5

()

Tretinoin

~

80

_~800 roc

5.1 Intermittent Regimen

CRABP

~E-~

600

60.s

400

40

c; ~

~ 200 20 ~ ~ ~ '0 O-'---'L.....J'---'---'---L--L---'--'---'---'-...LO 0:: ~

3

5 Day

8

15 after 7 days off

Fig. 2. Plasma clearance of tretinoin and amount of cyloplasmic retinoic acid binding proteins (CRABP) I and II measured in skin biopsy samples, both plotted as a function of days of tretinoin administration.[20[

• an increase in the binding and subsequent metabolism of drug within the cytosolic compartment of cells by upregulated CRABP. Furthermore, downregulation of CRABP, which accompanies the discontinuation of the drug, could result in less cytoplasmic binding of the drug and a potential increase in the effectiveness of tretinoin.

5. Modulation of Tretinoin Pharmacokinetics Several pharmacokinetic studies have shown that with long term administration plasma tretinoin concentrations decrease significantly over time. In APL patients, plasma drug exposure to tretinoin decreased significantly (40 to 90%) by day 7 of daily drug administration. In most patients the onset of the decrease in plasma drug concentration is relatively rapid and occurs within 2 to 7 days after beginning treatment. I [~, [<),4<).571 Different strategies aimed at preventing or overcoming the induced tretinoin resistance have been, and are being, planned. The principal strategies include intermittent administration, administration of pharmacological inhibitors of CYP oxidative enzymes, combination with IFNa and intravenous administration of liposome-encapsulated tretinoin.1 [71 Ultimately, however, the best solution may be the identification of retinoids which are equally potent but do not induce their own catabolism. © Adis International Limited. All rights reserved.

An intermittent regimen of tretinoin administration has a potential advantage over continuous administration as indicated by a recent preclinical pharmacology study of intravenous tretinoin.l 201 In that study tretinoin was given intravenously on 8 consecutive days to 4 rhesus monkeys; an additional dose was administered after 7 days without the drug. The plasma clearance of tretinoin increased (approximately doubled) during the 8-day period of drug administration, but returned to baseline concentrations after 7 days without it. Clinical studies carried out on patients with refractory cancer l571 and patients with HIV infection and Kaposi's sarcomal56,581 treated with tretinoin intermittently (7 days onl7 days off and 3 days on/4 days off) described a marked drop in plasma AUC ~ on the last day of the first cycle of therapy, but this was followed by a quick recovery of plasma tretinoin concentrations at the start of the following cycles of therapy (table II). We studied the pharmacokinetic profile of tretinoin (unpublished observations) in a group of 13 patients with Philadelphia chromosome-positive chronic myelogenous leukaemia (CML) in the chronic phase, they were treated with tretinoin 80 mg/m2/day in 2 divided doses for 7 consecutive days every other week (i.e. I week on and I week off equals 1 cycle). As shown in figure 3 and table II, tretinoin concentrations decreased significantly during the first week of drug administration, with a decrease in plasma AUC~ ranging from 26.5 to 92% on day '7 compared with day 1. Following I week without tretinoin, the mean plasma AUC ~ on day I of cycle 2 was not statistically different from the corresponding value observed on day I of cycle I. Our experience confirms the reversibility of the phenomenon of induced metabolism and the possibility of modulating tretinoin pharmacokinetics by changing the drug administration regimen, but also that: • the extent of the observed increase in plasma AUC~ after a week without the drug is unpredictable (plasma AUC ~ ranged between 22 and Clin. Phormacokinet. 1997 May; 32 (5)

396

Regazzi et al.

~ 300 Cl

5.2. 1 Metabolism Inhibition by Azole Antifungals

20

• Day 1 cycle 1 • Day 7 cycle 1 o Day 7 cycle 2

c 0

~

C 200 Q)

u c 0 u c '0 c 100

~ ctl

E (f)

ctl

a::

0 0

2

3 Time (h)

4

5

6

Fig. 3. Tretinoin plasma concentrations in 1 patient with Philadelphia chromosome-positive chronic myelogenous leukaemia treated with an intermittent regimen (7 days onn days off 1 cycle).

=

188% of the corresponding values observed on day I of the first week of therapy) • I week without tretinoin may not be enough in some patients to return the accelerated tretinoin metabolism to baseline values. However, the number of patients who underwent serial pharmacokinetic studies is small and further investigations are necessary to clarify the temporal relationship between drug-free interval and recovery of accelerated metabolism of tretinoin. Studies employing intermittent regimens with shorter drug-free intervals (3 days on/4 days off, or I day onll day off) are currently underway. In general, despite the fact that intermittent scheduling of tretinoin administration would seem to be a pharmacokinetic advantage, it is still not certain whether the higher plasma concentrations of tretinoin reached on the alternate regimen will be more effective than a long term daily regimen. Therefore, until the results of clinical trials planned to evaluate the role of intermittent tretinoin administration are available, we would caution against this therapeutic strategy. 5.2 Administration of Cytochrome P450 Inhibitors

Inhibition of tretinoin metabolism by treatment with CYP inhibitors might result in higher and more sustained cell levels of retinoic acid. © Adis Internationallirnited. All rights reserved.

Liarozole and ketoconazole inhibit the CYPdependent metabolism of retinoic acid. Inhibition of tretinoin metabolism by imidazole antimycotics was first observed in murine embryonal carcinoma cells.l 921 These observations have been extended to tretinoin metabolism in rat skin microsomes, hamster liver microsomes, human skin (epidermis), human skin equivalents and in vivo rats.l SO,83- 87 1These inhibitors compete with the substrate for binding to the haem iron of CYP. Ketoconazole and its analogues are inhibitors of both CYP and lipoxygenase enzymes. Thus, the observed effects may result not only from inhibiting tretinoin catabolism by CYP enzymes, but also by limiting necessary lipid peroxide cofactors which are generated by lipoxygenase. Liarozole enhanced the effects of retinoic acid in F9 teratocarcinomal cells, an effect shared with ketoconazole. 1931 In vivo ketoconazole and liarozole increased the plasma half-life of exogenously administered retinoic acid and enhanced its endogenous plasma concentrations. By inhibiting retinoic acid metabolism, liarozole may exert retinoid-mimetic effects and be used in the treatment of, for example, psoriasis and ichthyosis.1 94 ] The inhibition of the catabolism of tretinoin by liarozole also contributes to its antitumoural effect on several androgen dependent and, chiefly, androgen-independent R3327 Dunning prostate adenocarcinomas. 1951 Animal studies in which tretinoin was administered in combination with CYP enzyme inhibitors (e.g. ketoconazole and liarozole) showed a significant prolonging of the plasma half-life of tretinoin,18o, 87 1 thereby supporting the important contribution of the CYP enzyme system to the in vivo metabolic process of tretinoin. Two studies (similarly designed) by Rigas et al. I961 and Miller et al. 1971 reported that a single oral dose of ketoconazole 200 to 1200mg and liarozole 75 to 300mg given I hour before tretinoin attenuate the reduction in plasma tretinoin concentrations in Clin. Pharmacokinet. 1997 May: 32 (5)

Clinical Pharmacokinetics of Tretinoin

patients with solid tumours. Following continuous oral tretinoin treatment, the mean day 28 AUC of tretinoin was significantly lower than the mean AUC value on day 1 (Iiarozole: 504 vs 132 Ilg/L. h; ketoconazole: 467 vs 213 Ilg/L • h). This decline in plasma concentrations on day 28 was partially reversed by both liarozole and ketoconazole, which significantly increased the mean plasma tretinoin AUC on day 29 to 243 and 375 Ilg/L. h, respectively (fig. 4). Liarozole 300mg and ketoconazole 400mg were reported by the authors as the lowest doses to reliably produce this effect. [96,97] Even ifthese studies were not designed to demonstrate a sustained increase in plasma tretinoin concentrations over time, their results provide a rationale for combining ketoconazole or Iiarozole with tretinoin during prolonged treatment of patients with this retinoid,l 9S1 A randomised, crossover study[55] designed to evaluate the ability of prolonged administration of ketoconazole 600 mg/day to maintain plasma tretinoin concentrations in adults with solid tumours, showed that, after 2 weeks of coadministration with tretinoin, ketoconazole did not prevent the decrease of plasma tretinoin concentrations over time. The explanation for why long term ketoconazole did not maintain the concentrations of tretinoin in this study is unclear, but the lack of benefit does not appear to have resulted from inadequate ketoconazole doses (600mg is considered to be the maximum tolerated dose in patients with solid tumours) or from a lack of ketoconazole inhibition of CYP. In conclusion, while short term treatment with both ketoconazole and Iiarozole can attenuate the induced increase in catabolic rates, these combinations have not yet been shown to sustain high plasma concentrations on an extended basis. Future studies of the role of CYP in tretinoin metabolism are needed to identify agents with specific inhibitory effects on isoenzymes involved in tretinoin catabolism. 5.2.2 Metabolism Inhibition by Sulfonamides

It has been shown that CYP2C8, a human liver CYP isozyme, is capable of actively converting © Adis International Limited. All rights reserved.

397

600

£' 500

Tretinoin alone

Tretinoin alone T retinoin plus ketoconazole

o +--L---L-L---L._ 28

29

Day

Fig. 4. Plasma area under the concentration-time curve (AUC) of tretinoin increased significantly (p < 0.01) after administration (1 hour before tretinoin administration) of a single dose of ketoconazole l961 (doses ranging from 200 to 1200mg) or liarozole!971 (doses ranging from 75 to 300mg).

retinoic acid to polar metabolites, including 4hydroxy-retinoic and 4-oxo-retinoic acid.[SI] Recently evidence has been presented to show that tolbutamide is metabolised in humans primarily by the CYP isozymes 2C8, 2C9 and 2ClO.[99] The CYP2C family might playa major role in the metabolism of both tolbutamide and tretinoin. Since recent studies both in vitro and in vivo have shown marked competitive inhibition of tolbutamide hydroxylase activity by several sulfonamides, it is possible that these drugs have similar inhibitory effects on tretinoin metabolism) 100, 101 J We, in a preliminary study in animal models (piglets), have shown that tretinoin may induce tolbutamide metabolism and that sulphaphenazole may inhibit the metabolism of both tolbutamide and tretinoin. These results suggest a potential role of sulfonamides in modulating tretinoin pharmacokinetics (unpublished results). 5.2.3 Combination with Interferon-a.

Experimental and clinical data have demonstrated that the combination of interferons and tretinoin seems to be of particular interest: first for the possible additive anti proliferative and differentiative effects of 2 drugs; secondly, because interferons may playa role in limiting the intracellular tretinoin oxidation by downregulating the activity of the CYP enzymatic system,lI02, 103 1 elin. Phormac okinet. 1997 May; 32 (5)

398

Despite many studies in animals and, more recently, in humans, I104] the mechanism of the inhibition of hepatic oxidative drug metabolism by interferon is uncertain. It has been suggested that consequent to interferon administration there is an increased degradation of the haem moiety of CYP with production within the hepatocyte of an intermediate protein which could decrease the synthesis of certain CYP apoproteins. Besides this, interferons induce xanthine oxidase activity in the liver and oxygen radicals generated by this pathway may damage or destroy CYP apoproteins. The free radical scavenger a-tocopherol and the xanthine oxidase inhibitor allopurinol prevented interferonmediated loss of total CYP. Since only interferons with antiviral activity have shown the capability of depressing CYP, another possible explanation is that the biochemical pathways mediating the antiviral and anti tumour effects of interferons also mediate the depressive effect on hepatic CYP metabolism.IIOSI Lazzarino et al.I 1061 studied the effect of interferon on tretinoin pharmacokinetics in 2 patients with APL in molecular remission maintained by alternating 15 days of interferon 3 x 106 IU/m 2 every other day and 15 days of tretinoin 45 mg/m 21 day. After 9 and 22 months of maintenance therapy, mean AUC= values for tretinoin (determined on day 15 of tretinoin therapy) had decreased by 26.7% (of day 1 value) in patients who received combination therapy, whereas a decrease of 93.3% was seen in patients treated with tretinoin alone. The authors suggested a potential role for interferon in modulating tretinoin pharmacokinetics, providing a pharmacological basis to the clinical use of the interferon-tretinoin combination. This advantageous effect of interferon may be facilitated by a 15 days on/15 days off regimen of tretinoin administration which would allow, to some extent, the downregulation of catabolic mechanisms occurring physiologically during the withdrawal period. Previous studiesl1051 reported that the inhibitory effect of interferon is specific for constitutive forms of CYP enzymes (and non induced forms) 10 Adis International Limited. All rights reserved.

Regazzi et al.

and that interferons cause minimal differences in the plasma clearance of drugs, such as tretinoin, which are substrates of the 2C family of CYP. No effect of concurrent IFNa on the pharmacokinetics of tretinoin was seen in a phase I trial of IFNa + tretinoin. 1521 It should be taken into account, however, that interferon needs time to manifest its profound biological effects. Therefore it is possible that, with the relatively low interferon doses used, pharmacokinetic interactions with tretinoin appear only after several months of treatments. Prolonged interferon treatment may lead, through still unknown pathways, to downregulation of catabolic mechanisms involved in tretinoin clearance. The potential of interferon in modulating tretinoin pharmacokinetics may open new possibilities for tretinoin as a differentiating agent in long term therapies. Bailey et al.I 1071 studied concomitant administration of tretinoin on intermittent therapy and interferon subcutaneously daily in patients with HIVassociated Kaposi's sarcoma. Even if intermittent tretinoin administration was able to circumvent the low plasma exposure observed with continuous administration, this particular regimen yielded, at most, a very low response rate. 5.3 Liposome Formulation of Tretinoin

A new formulation of tretinoin encapsulated into liposomes (liposomal tretinoin) was recently developed to obtain a formulation which can be administered intravenously to provide potential pharmacological advantages over the oral formulation.l 1081 Animals (Lewis rats) treated long term with intravenous liposomal tretinoin metabolised retinoic acid to a lesser extent than animals treated with free tretinoin. After 7 weeks of continuous drug treatment, the concentrations of tretinoin in the blood of rats treated with free tretinoin decreased by 35%, whereas the mean blood tretinoin concentrations of rats treated with liposomal tretinoin did not changed significantly. It is suggested that lipid formulation bypasses the clearance mechanism which evolves in the liver of patients treated with the oral formulation,11091 Clin. Pharmacokinet. 1997 May: 32 (5)

Clinical Pharmacokinetics of Tretinoin

so that the liposomal formulation should not be subject to the same relapse rate which has been demonstrated in clinical trials with oral tretinoin administration. In addition, because liposomal encapsulation of tretinoin decreases the direct exposure of the drug during circulation to concentrations below the orally administered toxic dose, the adverse effects of liposomal tretinoin should be less severe.IIIOI The potential advantages of using liposomes as a delivery system include increased biological activity through specific targeting and decreased toxicity because of altered pharmacokinetics,!1111 The liposomally encapsulated form of treti no in, probably impractical for extended therapy because of the need for intravenous administration should provide a critical test of the pharmacological theory of acquired resistance.

6. Conclusions Tretinoin is a most challenging drug in a clinical pharmacokinetic perspective. It is generally agreed that its pharmacokinetics and utility are complicated by a number of known and unknown factors. Before the benefits of tretinoin can be exploited, it is important to learn how to prevent the induction of the accelerated metabolism, considered one of the major mechanisms of acquired clinical resistance to tretinoin. Despite the strategies which have sought to overcome the induced decrease in plasma drug concentrations, further studies are required to determine the best treatment approach. Of course, the main concern has to be for the patient. More clinical studies are emerging where the investigators are aware of the problems discussed above, and are trying to solve them in a multidisciplinary approach.

References

I. Cullum ME, Zile MH. Metabolism of all-trans-retinoic acid and all-tralls-retinyl acetate. J Bioi Chern 1985; 260: \0590-6 2. Degos L, Chomeienne C, Daniel MT, et al. Treatment of first relapse in acute promyelocytic leukemia with all-trwlS retinoic acid lletterl. Lancet 1990; 336 (8728): 1440-1 3. Castaigne S. Chomeienne C, Daniel MT, et al. Ali-trailS retinoic acid as differentiation therapy for acute promyelocytic leukemia. Blood 1990; 76: 1704-9

© Adis International Lirnited. All rights reserved.

399

4. Warrel RP, Frankel Jr RP, Miller Jr WH, et al. Differentiation therapy of acute promyelocytic leukemia with tretinoin (alltrails retinoic acid). N Engl J Med 1991; 324 (20): 1385-93 5. Fenaux P, Chastang C, Chomeienne C, et al. Tretinoin with chemotherapy in newly diagnosed acute pro myelocytic leukemia: European APL group Iletter). Lancet 1994; 343 (8904): \033 6. Kuzel T, Roenigk H, Samuelson E, et al. The treatment of mycosis fungoides with ali-trailS retinoic acid labstract). Blood 1993; 82: 570a 7. Gill PS, Espina BM, Mougdil T. et al. All-trans retinoic acid for the treatment of AIDS-related Kaposi's sarcoma: results of a pilot phase \I study. Leukemia 1994; 8 (3): S26-32 8. Lippman SM, Kavanagh JJ, Paredes-Espinoza M, et al. \3-cis retinoic acid plus interferon a-2a: highly active systemic therapy for squamous cell carcinoma of the skin. J Nat! Cancer Inst 1992; 84 (4): 241-5 9. Lippman SM, Parkinson DR, Itri LM, et al. 13-cis retinoic acid and interferon a-2a effective combination therapy for advanced squamous cell carcinoma of the skin. J Natl Cancer Inst 1992; 84 (4): 235-41 10. Athanasiadis I, Kies MS, Miller M, et al. Phase \I study of ali-trailS retinoic acid and interferon a-2a in patients with advanced non-small cell lung cancer. Clin Cancer Res 1995; I: 973 II. Anzano MA, Byers SW, Smith JM, et al. Prevention of breast cancer in the rat with 9-cis RA as a single agent and in combination with tamoxifen. Cancer Res 1994; 54 (17): 4614-7 12. Hill DL. Retinoids and cancer prevention. Annu Rev Nutr 1992; 12: 161-81 13. Huang ME. Yu-chen Y, Shu-rong C, et al. Use of ali-trailS retinoid acid in the treatment of acute promyelocytic leukemia. Blood 1988; 72: 567-72 14. Gillis JC, Goa KL. Tretinoin: a review of its pharmacodynamic and phannacokinetic properties and use in the management of acute promyelocytic leukemia. Drugs 1995; 50 (5): 897-923 15. Frankel SR, Eardley A, Heller G, et al. Ali-trailS retinoic acid for acute promyelocytic leukemia: results of the New York study. Ann Intern Med 1994; 120: 278-86 16. Warrell Jr RP, de-The H, Wang ZY, et al. Acute promyelocytic leukemia. N Engl J Med 1993; 329: 177-89 17. Warrell RP. Retinoid resistance in acute promyelocytic leukemia: new mechanisms, strategies and implications. Blood 1993; 82 (7): 1949-53 18. Lefebvre P, Thomas G, Gourmel B, et al. Pharmacokinetics of oral ail-trailS retinoic acid in patients with acute promyelocytic leukemia. Leukemia 1991; 5: 1054-8 19. Muindi J, Frankel SR, Miller WH, et al. Continuous treatment with ali-trailS retinoic acid results in a progressive decrease in plasma concentrations: implications for relapse and retinoid 'resistance' in acute promyelocytic leukemia. Blood 1992; 79: 299-303 20. Adamson PC, Boylan JF, Balis FM, et al. Time course of induction of metabolism of all-tralls-retinoic acid and the up-regulation of cellular retinoic acid-binding protein. Cancer Research 1993; 53: 472-6 21. Pemrick SM, Lucas DA, Grippo JE The retinoid receptors. Leukemia 1994; 8: 1797-806 22. Dermine S, Grignani F, Monica C. et al. Occurrence of resistance to retinoic acid in the acute promyelocytic leukemia cell line NB4 is associated with altered expression of the PMLlRAR-a protein. Blood 1993; 82: 1573-7

Clin. Pharmacokinet. 1997 May; 32 (5)

400

23. De Ruyter MGM, De Leenheer AP. Retinoic acid, an endogenous compound of human blood: unequivocal demonstration of endogenous retinoic acid in normal physiological conditions. Anal Biochem 1979; 98: 402-9 24. McCormick AM, Napoli JM, DeLuca HE High pressure liquid chromatography of vitamin A metabolites and analogs. Methods Enzymol 1980; 67: 220-33 25. Rissler K, Friedrich G. Determination of the retinoid (E)I ,2,3,4,-tetrahydro-l, I ,4,4-tetramethyl-6-( I methyl-2-phenylethenyl)naphtalene and its phenolic metabolite in human plasma by reversed-phase high performance liquid chromatography. J Chromatogr 1991; 565: 375-82 26. Noy N. The ionization behaviour of retinoic acid in aqueous environments and bound to serum albumin. Biochim Biophys Acta 1992; 1106: 152-8 27. Noy N. The ionization behaviour of retinoic acid in lipid bilayers and in membranes. Biochim Biophys Acta 1992; 1106: 159-64 28. Boylan JF, Gudas LJ. Overexpression of the cellular retinoic acid binding protein-I (CRABP-I) results in a reduction in differentiation specific gene expression in F9 teratocarcinoma cells. J Cell Bioi 1991; 112: 965-70 29. Boylan JF, Rugas LJ. The level of CRABP I expression influences the amounts and types of all-trans retinoic acid metabolites in F9 teratocarcinoma stem cells. J Bioi Chern 1992; 267: 21486-91 30. Kizaki M, Ikeda Y, Tanosaki R, et al. Effects of novel retinoic acid compound, 9-cis-retinoic acid, on proliferation, differentiation and expression of retinoic acid receptor-a and retinoid X receptor-a RNA by HL-60 cells. Blood 1993; 12 (15): 3592-9 31. Goodman DS, Blaner WS. Biosynthesis, absorption and hepatic metabolism of retinol. In: Sporn MB, Roberts AB, Goodman DS, editors. The retinoids. Vol. 2. New York: Raven Press, 1984: 1-40 32. Fidge NH, Shiratori T, Ganguly J, et al. Pathway of absorption of retinal and retinoic acid in the rat. J Lipid Res 1968; 9: 103-9 33. Smith JE, Milch PO, Muto Y, et al. The plasma transport and metabolism of retinoic acid in the rat. Biochem J 1973; 132: 821-7 34. Kalin JR, Starling ME, Hill DE. Disposition of all-trans retinoic acid in mice following oral doses. Drug Metab Disp 1981; 9: 196-201 35. EckhoffC, Collins MD, Nau H. Human plasma ali-trans, 13 cis and 13-cis-4-oxo-retinoic acid profiles during subchronic vitamin A supplementation: comparison to retinol and retinyl ester plasma levels. J Nutr 1991; 121: 1016-25 36. Folman Y, Russell RM, Tang GW, et al. Rabbits fcd on beta carotene have higher serum levels of all-trans retinoic acid than those receiving no beta carotene. Br J Nutr 1989; 62: 195-201 37. Napoli JL, Race KR. Biogenesis of retinoic acid from ~-caro­ tene. J Bioi Chern 1988; 263: 17372-7 38. Posch KC, Boerman MHEM, Burns RD, et al. Holo-cellular retinol-binding proteins as a substrate for microsomal retinal synthesis. Biochemistry 1991; 30: 6224-30 39. Sporn MB, Roberts AB, Goodman DS. The retinoids: biology, chemistry and medicine. 2nd ed. New York: Raven Press Ltd, 1994 40. Eriksson U, Hansson E, Nordlinder H, et al. Quantitation and tissue localization of the cellular retinoic acid binding protein. J Cell Physiol 1987; 133: 482-9

© Adis International Limited. All rights reserved.

Regazzi et al.

41. Saari JC, Futterman S. Separable binding proteins for retinoic acid and retinol in bovine retina. J Bioi Chern 1976; 253: 6432-6 42. Sundelin J, Das SR, Eriksson U, et al. The primary structure of bovine cellular retinoic acid-binding proteins. J Bioi Chern 1985;260: 6488-93 43. Chytil F, Page OL, Ong DE. Presence of cellular retinol and retinoic acid binding proteins in human huterus. Int J Vitam Nutr Res 1975; 45: 293-8 44. Elder JT, Astrom A, Pettersson U, et al. Differential regulation of retinoic acid receptors and binding proteins in human skin. J Invest Dermatol 1992; 98: 673-9 45. Petkovich M, Bran NJ, Krust A, et al. A human RA receptor which belongs to the family of nuclear receptors. Nature 1987; 330: 444-50 46. Mangelsdorf OJ, Ong ES, Dyck JA, et al. Nuclear receptor that identifies a novel RA response pathway. Nature 1990; 345: 224-9 47. Nepgal S, Saunders M, Kastner P, et al. Promoter context and response element-dependent specificity of the transcriptional acti vation and modulating functions of RA receptors. Cell 1992; 70: 1007-19 48. Blumberg B, Mangelsdorf DJ, Dyck JA, et al. Multiple retinoidresponsive receptors in a single cell: families of retinoid 'X' receptors and retinoic acid receptors in the xenopus egg. Proc Natl Acad Sci USA 1992; 89: 2321-5 49. Muindi JRF, Frankel SR, Huselton C, et al. Clinical pharmacology of oral all-trans-retinoic acid in patients with acute promyelocytic leukemia. Cancer Res 1992; 52: 2138-42 50. Muindi JRF, Young CW, Warrell RP. Clinical pharmacology of all-trans-retinoic acid. Leukemia 1994; 8: 1807-12 51. Smith MA, Adamson PC, Balis FM, et al. Phase I and pharmacokinetic evaluation of all-trans retinoic acid in pediatric patients with cancer. J Clin Oncol 1992; 10 (II) 1666-73 52. Lee JS, Newman RA, Lippman SM, et al. Phase I evaluation of all-trans retinoic acid in adults with solid tumors. J Clin Oncol 1993; II (85): 959-66 53. Castaigne S, Lefebvre P, Chomienne C, et al. Effectiveness and pharmacokinetics of low-dose all-trans retinoic acid (25 mg/m 2) in acute promyelocytic leukemia. Blood 1993; 82: 3560-3 54. Smith DC, Jacob HE, Lotze MT, et al. A phase I trial of interferon-a 2a (IFN-a) and all-trans retinoic acid (ATRA): a pharmaeokinetie assessment. Proc Am Soc Clin Oneol 1994; 13: 134 55. Lee JS, Newman RA; Lippman SM, et al. Phase I evaluation of all-trans retinoie acid with and without ketokonazole in adults with solid tumors. J Clin Oneol 1995; 13 (6): 1501-8 56. Adamson PC, Bailey J, Pluda J, et al. Pharmacokinetics of alltrans-retinoic acid administered on an intermittent schedule. J Clin Oncol 1995; 13: 1238-40 57. Adamson PC, Reaman G, Feusner 1. Pediatric phase I trial and pharmacokinetic study of all-trans retinoic (ATRA) acid administered on an intermittent schedule. Proc Am Soc Clin Oncol 1995; 14: 170 58. Toma S, lacona I, Palumbo R, et al. Modulation of all-trans retinoic acid administered on intermittent schedule by a-interferon 2a in a patient with AIDS-related Kaposi's sarcoma. AIDS. 1996; 10 (9): 1049-50 59. EI Mansouri S, Tod M, Leclerq M, et al. Time- and dose-dependent kinetics of all-trans retinoic acid in rats after oral or intravenous administration. Drug Metab Dispos 1995; 23 (2): 227-31

Clin. Pharmacokinet. 1997 May: 32 (5)

Clinical Pharmacokinetics of Tretinoin

60. Adamson PC, Balis FM, Smith M A, et al. Dose-dependent pharmacokinetics of all-trans retinoic acid. J Natl Cancer Inst 1992; 84: 1332-5 61. Adamson Pc. Pharmacokinetics of ali-trails retinoic acid: clinical implications in acute promyelocytic leukemia. Semin Hematol1994; 31 (5): 14-7 62. Data on file, F. Hoffmann-La Roche, Basel, Switzerland 63. Skare KL, Deluca HF. Biliary metabolites of all-trans-retinoic acid in the rat. Arch Biochem Biophys 1983; 224: 13-8 64. Zachman RD, Dunagin Jr PE, Olson JA. Formation and enterohepatic circulation of metabolites of retinol and retinoic acid in bile duct-cannulated rats. J Lipid Res 1966; 7: 3-9 65. Geison RL, Johnson BC. Studies on the in vivo metabolism of retinoic acid in the rat. Lipids 1970; 5: 371-8 66. Ito Y, Zile M, De Luca HF, et al. Metabolism of retinoic acid in vitamin A-deficient rats. Biochim Biophys Acta 1974; 369: 338-50 67. Adamson PC , Pitot HC, Balis F, et al. Variability in the oral bioavailability of all-trans retinoic acid. J Natl Cancer Inst 1993; 85: 993-6 68. Barua AB, Batres RO, Olson JA. Characterization of retinyl ~-glucuronide in human blood. Am J Clin Nutr 1989; 50: 370-4 69. Barua AB, Gunning DG, Olson JA. Metabolism in vivo of alltrans-I 11-1HJ retinoic acid after an oral dose in rats. Biochem J 1991; 277: 527-31 70. Blaner WS, Hussain MM, Talmage DA, et al. Retinoids are actively metabolized in rabbit bone marrow. FASEB J 1991; 5: A717 71. Kakizuka A, Miller Jr WH, Umesono K, et al. Chromosomal translocation t(l5; 17) in human acute promyelocytic leukemia fuses RARa with a novel putative transcriptional factor, PML. Cell 1991; 68: 663-74 72. de-The H, Lavau C, Marchio A, et al. The PML-RARa fusion mRNA generated by the t(l5; 17) translocation in acute promyelocytic leukemia encodes a functionally altered RAR. Cell 1991; 66: 675-84 73. Sandberg JA, Eckhoff C, Nau H, et al. Pharmacokinetics of 13-cis, ali-trailS, 13-cis-4-oxo, and all-trans retinoic acid after intravenous administration in the cynomolgi us monkeys. Drug Metab Disp 1994; 22 (I): 154-60 74. Fiorella PO, Napoli JL. Microsomal retinoic acid metabolism. J Bioi Chem 1994; 269: 10538-44 75. Roberts AB, Nichols MD, Newton DL, et al. In vitro metabolism of retinoic acid in hamster intestine and liver. J Bioi Chem 1979; 254: 6296-302 76. Frolik CA, Roberts AB, Tavela TE, et al. Isolation and identification of 4-hydroxy and 4-oxo retinoic acid: in vitro metabolites of ali-trailS retinoic acid in hamster trachea and liver. Biochemistry 1979; 18: 2092-7 77. Frolik CA, Roller PP, Roberts AB, et al. In vitro and in vivo metabolism of all-trans retinoic acid and 13-cis retinoic acid in hamster. J Bioi Chem 1980; 225: 8057-62 78. Roberts AB, Lamb LC , Sporn MB. Metabolism of all-transretinoic acid in hamster liver microsomes: oxidation of 4hydroxy- to 4-keto-retinoic acid. Arch Biochem Biophys 1980; 199: 374-83 79. Martini R, Murray M. Participation of 3A enzymes in rat hepatic microsomal retinoic acid 4-hydroxylation. Arch Biochem Biophys 1993; 303: 57-66 80. Van Wauwe J, Van Nyen G, Coene MC, et al. Liarozole, an inhibitor of retinoic acid metabolism exert retinoid-mimetic effects in vivo. J Pharmacol Exp Ther 1992; 261: 773-9

© Adis International Limited. All rights reserved.

401

81. Roberts ES, Vaz ADN, Coon MJ. Role of isozymes of rabbit microsomal cytochrome P-450 in the metabolism of retinoic acid, retinol and retinal. Mol Pharmacol 1992; 41: 427-33 82. Leo MA, Lasker JM, Raucy JL, et al. Metabolism of retinol and retinoic acid by human liver cytochrome P-450IlC8. Arch Biochem Biophys 1989; 269: 305-12 83. Cornie M, Delva L, Castaigne S, et al. In vitro ali-trailS retinoic acid sensitivity and cellular retinoic acid binding protein levels in relapse leukemic cells after remission induction by ATRA in acute promyelocytic leukemia. Leukemia 1994; 8: 914-7 84. Vanden Bossche H, Willemsens G, Janssen PAJ. Cytochrome P-450-dependent metabolism of retinoic acid in rat skin microsomes: inhibition by ketoconazole. Skin Pharmacol 1988; I: 176-85 85. Vanden Bossche H, Willemsens G. Retinoic acid and cytochrome P-450. In: Saurat JH, editor. Retinoids 10 years on. Basel: Karger, 1991: 79-88 86. Van Wauwee JP, Coene MC, Goossens J, et al. Ketoconazole inhibits the in vitro and in vivo metabolism of all-trans retinoic acid. J Pharmacol Exp Ther 1988; 245: 718-22 87. Van Wauwee JP, Coene MC, Goossens J, et al. Effects of cytochrome P-450 inhibitors on the in vivo metabolism of alltrans retinoic-acid in rats. J Pharmacol Exp Ther 1990; 252: 365-9 88. Adamson Pc. Clinical and pharmacokinetic studies of all-trans retinoic acid in pediatric patients with cancer. Leukemia 1994; 8 ( II): 1813-6 89. Fiorella PO, Napoli JL. Expression of cellular retinoic acid binding protein (CRABP) in Escherichia coli: characterization and evidence that holo CRABP is a substrate in retinoic acid metabolism. J Bioi Chem 1991; 266: 16572-9 90. Tavakkol A, Griffith CEM, Keane KM, et al. Cellular localization of mRNA for cellular retinoic acid-binding protein II and nuclear retinoic acid receptor in retinoic-acid treated human skin. J Invest Dermatol 1992; 99: 146-50 91. Hirschel-Scholz S, Siegenthaler G, Saurat JH. Ligand specific and non-specific in vivo modulation of human epidermal cellular retinoic acid binding proteins (CRABP). Eur J Clin Invest 1989; 19: 220-7 92. Williams JB, Napoli JL. Metabolism ofretinoic acid and retinol during differentiation of F9 embryonal carcinoma cells. Biochem Pharmacol 1985; 36: 1386-8 93. De Coster R, Wouters W, Van Ginckel R, et al. Experimental studies with liarozole (R 75251): an antitumoral agent which inhibits retinoic acid breakdown. J Steroid Biochem Mol Bioi 1992; 43: 197-201 94. Goldfarb MT, Ellis CN, Voorhees JJ. Retinoids in dermatology. Mayo Clin Proc 1987; 62: 1161-4 95. Dijkman GA, Van Moorselaar RJA, Van Ginkel R, et al. Antitumoral effects of Iiarozole in androgen-dependent and independent R3327-Dunning prostate adenocarcinomes. J Urol 1994; 151: 217-22 96. Rigas JR, Francis PA, Muindi JRF, et al. Constitutive variability in the pharmacokinetics of the natural retinoid, all-transretinoic acid, and its modulation by ketoconazole. J Nat! Cancer Inst 1993; 85: 1921-6 97. Miller VA, Rigas JR, Muindi JR, et al. Modulation of all-trans retinoic acid pharmacokinetics by liarozole. Cancer Chemother Pharmacol 1994; 34 (6): 522-6 98. Westarp ME, Westarp MP, Bruynseels J, et al. Oralliarozole as a catabolic inhibitor potently increases retinoic acid in vivo: first experience from an ongoing therapeutic trial in highly malignant primary brain tumors. Onkologie 1993; 16: 22-5

Clin. Pharmacokinet. 1997 May: 32 (5)

Regazzi et al.

402

99. Reiling MV, Aoyama T, Gonzalez FJ, et al. Tolbutamide and mephenytoin hydroxylation by human cytochrome P-450s in the CYP2C subfamily. J Pharmacol Exp Ther 1990; 152: 442-7 100. Veronese ME, Miners JO, Randles 0, et al. Validation of the tolbutamide metabolic ratio for population screening using sulphaphenazole to produce model phenotyping poor metabolisers. Clin Pharmacol Ther 1990; 47: 403-11 101. Pond SM, Birkett OJ, Wade ON. Mechanism of inhibition of tolbutamide metabolism: phenylbutazone, oxyphenbutazone, sulfaphenazole. Clin Pharmacol Ther 1977; 22: 573-9 102. Toma S, Monteghirfo S, Tasso P, et al. Antiproliferative and synergistic effect of interferon a-2a, retinoids and their association in established human cancer cell lines. Cancer Lett 1994; 82: 209-16 103. Okuno H, Kitao Y, Takasu M, et al. Depression of drug metabolizing activity in the human liver by interferon a. Eur J Clin Pharmacol 1990; 39: 365-7 104. Moor DM, Kalvakolanu DV, Lippman SM, et al. Retinoic acid and interferon in human cancer: mechanistic and clinical studies. Semin Hematol 1994; 31: 31-7 105. Israel BC, Blouin RA, Mcintyre W, et al. Effect of interferon a monotherapy on hepatic drug metabolism in cancer patients. Br J Clin Pharmacol 1993; 36: 229-35 106. Lazzarino M, Corso A, Regazzi MB, et al. Modulation of alltrails retinoic acid pharmacokinetics in acute promyelocytic

© Adis International Umited. All rights reserved.

107.

108.

109. 110.

III.

leukemia by prolonged interferon a therapy. Br J Haematol 1995; 90: 928-30 Bailey J, Pluda JM, Foli A, et al. Phase 1111 study of intermittent all-tralls-retinoic-acid alone and in combination with interferon alpha-2a, in patients with epidemic Kaposi's sarcoma. J Clin Oncol 1995; 13: 1966-74 Mehta K, Sadeghi T, McQueen T, et al. Liposome encapsulation circumvents the hepatic clearance mechanisms of ali-trails retinoic acid. Leukemia Res 1994; 18 (8): 587-96 Gray A, Morgan 1. Liposomes in haematology. Blood Rev 1991; 5: 258-72 Phillips NC, Tsoukas C. Liposomal encapsulation of azidothymidine results in decreased hematopoietic toxicity and enhanced activity against murine acquired immunodeficiency syndrome. Blood 1992; 79: 1137-43 Drach J, Lopez-Berestein G, McQueen T, et al. Induction of differentiation in myeloid leukemia cell lines and acute promyelocytic leukemia cells by Iiposomal ali-trails retinoic acid. Cancer Res 1993; 53: 2100-4

Correspondence and reprints: Dr Mario B. Regazzi, Department of Pharmacology, IRCCS-S. Matteo Hospital, P Ie Golgi 2, 27100 Pavia, Italy.

Clin. Pharmacokinet. 1997 May: 32 (5)