Clin Pharmacokinet 2004; 43 (5): 311-327 0312-5963/04/0099-0311/$31.00/0

REVIEW ARTICLE

© 2004 Adis Data Information BV. All rights reserved.

Clinical Pharmacokinetics of Thalidomide Steve K. Teo,1 Wayne A. Colburn,2 William G. Tracewell,3 Karin A. Kook,4 David I. Stirling,1 Markian S. Jaworsky,1 Michael A. Scheffler,1 Steve D. Thomas1 and Oscar L. Laskin1 1 2 3 4

Celgene Corporation, Warren, New Jersey, USA MDS Pharma Services, Phoenix, Arizona, USA MDS Pharma Services, Lincoln, Nebraska, USA Salamandra LLC, Chevy Chase, Maryland, USA

Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 1. Physical and Chemical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 2. Plasma Thalidomide Assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 3. Chiral Inversion and Protein Binding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314 4. Dosage Regimens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314 5. Metabolism and Hydrolysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314 6. Single- and Multiple-Dose Pharmacokinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 6.1 Formulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 6.2 Bioavailability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 6.3 Single-Dose Studies in Healthy and Patient Populations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 6.4 Multiple-Dose Studies in Healthy and Patient Populations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 6.5 Dose Proportionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 6.6 Effects of Age, Sex, Smoking and Food . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 6.7 Elimination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 6.8 Adverse Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322 7. Concentration-Response Relationship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322 8. Pharmacokinetic Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 8.1 Oral Contraceptives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 8.2 Warfarin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324 8.3 Other Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 9. Hepatic and Renal Dysfunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325 10. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325

Abstract

Thalidomide is a racemic glutamic acid derivative approved in the US for erythema nodosum leprosum, a complication of leprosy. In addition, its use in various inflammatory and oncologic conditions is being investigated. Thalidomide interconverts between the (R)- and (S)-enantiomers in plasma, with protein binding of 55% and 65%, respectively. More than 90% of the absorbed drug is excreted in the urine and faeces within 48 hours. Thalidomide is minimally metabolised by the liver, but is spontaneously hydrolysed into numerous renally excreted products. After a single oral dose of thalidomide 200mg (as the US-approved capsule formulation) in healthy volunteers, absorption is slow and extensive, resulting in a peak concentration (Cmax) of 1–2 mg/L at 3–4 hours after administration, absorp-

312

Teo et al.

tion lag time of 30 minutes, total exposure (AUCβ ) of 18 mg ± h/L, apparent elimination half-life of 6 hours and apparent systemic clearance of 10 L/h. Thalidomide pharmacokinetics are best described by a one-compartment model with first-order absorption and elimination. Because of the low solubility of the drug in the gastrointestinal tract, thalidomide exhibits absorption rate-limited pharmacokinetics (the ‘flip-flop’ phenomenon), with its elimination rate being faster than its absorption rate. The apparent elimination half-life of 6 hours therefore represents absorption, not elimination. The ‘true’ apparent volume of distribution was estimated to be 16L by use of the faster elimination-rate half-life. Multiple doses of thalidomide 200 mg/day over 21 days cause no change in the pharmacokinetics, with a steady-state Cmax (Cssmax) of 1.2 mg/L. Simulation of 400 and 800 mg/day also shows no accumulation, with Cssmax of 3.5 and 6.0 mg/ L, respectively. Multiple-dose studies in cancer patients show pharmacokinetics comparable with those in healthy populations at similar dosages. Thalidomide exhibits a dose-proportional increase in AUC at doses from 50 to 400mg. Because of the low solubility of thalidomide, Cmax is less than proportional to dose, and tmax is prolonged with increasing dose. Age, sex and smoking have no effect on the pharmacokinetics of thalidomide, and the effect of food is minimal. Thalidomide does not alter the pharmacokinetics of oral contraceptives, and is also unlikely to interact with warfarin and grapefruit juice. Since thalidomide is mainly hydrolysed and passively excreted, its pharmacokinetics are not expected to change in patients with impaired liver or kidney function.

Thalidomide was first synthesised in 1954 by the Chemie Grunenthal company of Germany. It is a chiral compound and is prescribed as the equimolar racemate consisting of (+)-(R)- and (–)-(S)-thalidomide. It was used in Europe as a non-barbiturate hypnosedative and antiemetic for morning sickness. The sedative effect of thalidomide is through a different mechanism than that of the barbiturates. Unlike barbiturates, thalidomide does not depress CNS neuronal function via enhancement of GABA-mediated neurotransmission, but rather is thought to activate sleep centres in the forebrain.[1] It was therefore believed to be a ‘safe’ drug, with little CNS and respiratory depression or muscle incoordination.[2] In fact, no mortality from overdose or attempted suicide has ever been recorded, and the highest documented overdose is of 14g with the patient making a full recovery.[3] Overdose results in moderate deep sleep without need for any supporting therapy upon awakening. Thalidomide was withdrawn in 1961 when its teratogenic effect in humans surfaced. By then, 5000–6000 babies from 46 countries had been born © 2004 Adis Data Information BV. All rights reserved.

with various external and internal deformities, including phocomelia, deafness, facial and oculomotor paralysis, and cardiac, uterine and vaginal malformations.[4] Thalidomide produces variable teratogenic responses in rabbits. Certain primates produce live foetuses with characteristic defects. Rodents are somewhat resistant, responding with increased fetal resorption.[5,6] If a non-rodent species such as rabbit had also been used in the initial testing, the teratogenicity of thalidomide would have been evident and it would not have been approved in Europe. The US was spared the tragedy through the diligence of the FDA reviewer Frances Kelsey.[7] In 1960, the Richardson-Merrell company applied for approval of thalidomide as a sedative. Approval was withheld, pending response to Dr Kelsey’s requests for more data on the peripheral neuropathy reported in some European patients. In 1965, the Israeli dermatologist Jacob Sheskin used thalidomide as a sedative in leprosy patients who had trouble sleeping due to painful inflammatory skin lesions called erythema nodosum leprosum (ENL). He serendipitously discovered that the leClin Pharmacokinet 2004; 43 (5)

Thalidomide

sions cleared rapidly and, following confirmation of its efficacy in controlled trials, thalidomide has been the drug of choice for ENL.[8] Even though the US did not approve thalidomide in the 1960s, it has been available on an investigational basis for the treatment of ENL, cancer and various inflammatory and immune diseases such as aphthous ulcers, graftversus-host disease, Beh¸ cet’s syndrome, rheumatoid arthritis and cutaneous lupus from the 1960s until its approval for ENL in 1998.[9,10] The proprietary capsule formulation of thalidomide (Thalomid® 1, Celgene, Warren, NJ, US) is the only form of thalidomide approved in the US. Since 1998, over 150 clinical trials in various inflammatory and oncologic conditions have been initiated. Thalidomide has recently shown promising antitumour activity in refractory multiple myeloma, renal carcinoma and glioblastoma multiforme, and almost eliminates the gastrointestinal adverse effects of irinotecan chemotherapy in metastatic colorectal cancer.[11-14] The anti-inflammatory and immunomodulatory activities of thalidomide are thought to be due to degradation of mRNA encoding tumour necrosis factor • (TNF-• ) in monocytes.[15] Thalidomide has also been shown to be anti-angiogenic in the rabbit cornea micropocket assay.[16] The combination of anti-inflammatory and anti-angiogenic actions make thalidomide a novel therapeutic agent with significant potential in treating a wide variety of diseases. Until recently there were no pharmacokinetic studies of thalidomide in humans, since its use was limited and restricted. Over the years, encouraging data from investigative studies in various diseases have given rise to single- and multiple-dose pharmacokinetic studies over a dose range of 100–1200mg.[17-27] The clinical pharmacology of thalidomide and its enantiomers has previously been presented.[28] The current review expands on this and describes the oral pharmacokinetics of various formulations of thalidomide in healthy and patient populations, with emphasis on the currently marketed capsule formulation. We also describe the absorption rate-limited pharmacokinetics, food effects and recent findings on potential drug interactions and renal dysfunction. The pharmacokinetics of an 1

313

experimental intravenous formulation are also briefly described. 1. Physical and Chemical Properties Thalidomide [(τ )-• -(N-phthalimido)glutarimide; C13H10N2O4] is a neutral racemic compound derived from glutamic acid. It consists of equimolar amounts of (+)-(R)- and (–)-(S)-enantiomers, is a stable white crystalline powder, has a molecular weight of 258.2 and a melting point of 275–278≤C (figure 1). Thalidomide is sparingly soluble in water (<0.1 g/L) with an n-octanol/water partition coefficient of 5.0[10]. It is stable in the solid form but spontaneously hydrolyses in solution at pH 6.0 or higher to produce at least 12 hydrolysis products (Teo S, unpublished data). It is stabilised in aqueous media through acidification.[29] 2. Plasma Thalidomide Assay Consistent with the authors’ experience, studies showed that thalidomide in unbuffered plasma has a room temperature half-life of approximately 7 hours, compared with 44 days for buffered plasma. All plasma samples containing thalidomide should therefore be buffered during sample processing and storage.[17-27,29,30] Therefore, in the Celgene studies with Thalomid®, plasma samples were stabilised with Sorensen’s citrate buffer at pH 1.5 prior to work-up and assay by high performance liquid chromatography (HPLC) or liquid chromatography-tandem mass spectrometry (LC-MS/MS). In our HPLC assay, thalidomide was extracted from buffered plasma with ethyl acetate and the reconstituted extracts were then injected into a reverse-phase HPLC system followed by UV detection at 218nm.[22] Standard curves were linear from 0.1 to 5 mg/L. Accuracy was 99–108%, precision of quality control samples was <4.5%, and the limit of detection was O N

O *

H N O

O Fig. 1. Structure of thalidomide. The asterisk denotes the chiral centre.

The use of trade names is for product identification purposes only and does not imply endorsement.

© 2004 Adis Data Information BV. All rights reserved.

Clin Pharmacokinet 2004; 43 (5)

314

Teo et al.

0.1 mg/L. In the highly sensitive LC-MS/MS assay, sample workup was similar with linearity demonstrable from 2 to 250 g/L, a precision of <12.3% and a limit of detection of 2 g/L.[29] Neither assays distinguished between the enantiomers, and total thalidomide concentration was determined. Some studies with other formulations did not include buffering of the plasma samples.[17,27] One study found that acidic buffering was not needed to reduce spontaneous hydrolysis and that thalidomide was stable during sample processing and storage.[17] 3. Chiral Inversion and Protein Binding Numerous studies have been performed on the (S)- and (R)-enantiomers of thalidomide. The enantiomers have different pharmacokinetic and pharmacodynamic properties.[31-35] The teratogenic effect of thalidomide is thought to be due to the (S)-enantiomer, with the (R)-enantiomer causing the sedation.[31-33] The immunomodulating effects are thought to be due to the (S)-enantiomer.[36] Development of the (R)-enantiomer as a drug is hampered by chiral interconversion in humans, catalysed by human serum albumin, producing both enantiomers.[34] Production of the (R)-enantiomer was found to be 1.6 times greater when compared using the area under the concentration-time curve (AUC), suggesting greater elimination of the (S)-enantiomer.[35] Protein binding in human plasma has never been determined for racemic thalidomide, however binding of the (R)- and (S)-enantiomers to human serum albumin was estimated to be 55 and 65%, respectively. Low binding was also seen in red blood cells.[34] Thalidomide also binds to another plasma protein, • 1-acid glycoprotein.[37] This binding is thought to modulate the immunological activity of thalidomide via TNF-• . 4. Dosage Regimens Due to its sedative property, thalidomide is usually taken once a day at bedtime. For ENL, the initial dosage is usually 200 mg/day increasing to 400 mg/ day.[38] The dosage is then decreased over time to give an acceptable adverse effect profile. Over the long term, a maintenance dosage of 50–200 mg/day is usually enough for the control of ENL skin lesions. Thalidomide is being assessed experimentally © 2004 Adis Data Information BV. All rights reserved.

for oncologic indications such as multiple myeloma, where the starting dosage is 200 mg/day increasing to 800 mg/day with maintenance at 200–400 mg/ day.[11] Additional investigational use in inflammatory indications such as Crohn’s disease also starts at 200 mg/day, with maintenance at 100–200 mg/ day.[39] 5. Metabolism and Hydrolysis Numerous studies have been performed on thalidomide metabolism and hydrolysis.[40-47] An in vitro study found that thalidomide metabolism required liver microsomes, an NADPH-regenerating system and molecular oxygen. In these conditions, uninduced microsomes from mice, rats and dogs were able to activate thalidomide, whereas Arochlor 1254-induced microsomes from rats were not.[40] Studies in rats showed an increase in the hepatic cytochrome P450 (CYP) content after oral administration at 3 mg/kg for 10 days.[41] The involvement of hepatic metabolic enzymes is indirectly supported by longer term studies in mice and rats given up to 3000 mg/kg orally for 13 weeks. These animals had 4 and 3% increases in the liver to bodyweight ratios respectively. It is not known for certain if this is due to minor induction of hepatic CYP isozymes.[42] In vitro hepatic metabolism studies in rats and rabbits have shown the production of the 5-hydroxy compound as the major hepatic thalidomide metabolite.[43] Recent in vitro studies have shown that the CYP2C subfamily mediates the 5and 5 -hydroxylation of thalidomide, with CYP2C19 involved in humans and CYP2C6 and 2C11 in rats. The 4- and N-hydroxy metabolites were not detected.[44] Human studies, however, show that metabolic production of the 5-hydroxy metabolite is very low and that thalidomide is minimally metabolised by the CYP system (figure 2).[45,46] Thalidomide is mainly broken down through nonenzymatic hydrolytic cleavage. This hydrolysis is spontaneous and occurs in aqueous solutions at physiological pH, with the products passively eliminated in the urine.[47-49] The extent of hydrolysis has been estimated to be 8 and 80% at 1 and 24 hours, respectively. Hydrolysis involves cleavage of all four amide bonds, thereby opening the glutarimide and phthalimide rings, producing 12 products, 11 of Clin Pharmacokinet 2004; 43 (5)

Thalidomide

315

OH

O

O

N

H N

*

O

O 4-OH-thalidomide

HO

O

O N

H N

*

O

O O

O N

5-OH-thalidomide

H N

*

O

O O

O N

H N

*

O HO

Thalidomide

O O

OH

O

5'-OH-thalidomide

N N

*

O

O N-OH-thalidomide Fig. 2. Proposed hepatic cytochrome P450-mediated metabolites of thalidomide. Hepatic metabolism is minimal (adapted from Teo et al.,[46] with permission).

which are chiral. These chiral products are further broken down to yield numerous optically active compounds.[47] As far as we know, no studies have been performed on the activity of the chiral hydrolysis products, presumably because of their aqueous instability and subsequent breakdown into numerous chiral byproducts. 6. Single- and Multiple-Dose Pharmacokinetics 6.1 Formulations

Prior to 1998, different capsule and tablet formulations of thalidomide were available for investigational use in the US. The Thalomid® capsule formulation (Celgene), first marketed in 1998, is the only © 2004 Adis Data Information BV. All rights reserved.

approved formulation. Different pharmacokinetics have been observed with the various formulations. The current review will emphasise the pharmacokinetics of the Thalomid® capsule formulation because the study conditions are best understood by us. A comparison will be made with data from a singledose intravenous study where drug delivery and disposition are well described. Published studies in some patient populations used formulations other than Thalomid®, and the physical and chemical properties of these other formulations are unknown to us. They probably contained different incipients, excipients and had different particle sizes and morphic forms of thalidomide, thereby producing different dissolution characteristics. This could have given rise to differences in some pharmacokinetic Clin Pharmacokinet 2004; 43 (5)

316

Teo et al.

parameters compared with Thalomid® (table I and table II). 6.2 Bioavailability

The low solubility of thalidomide has prevented development of a commercial intravenous formulation. In a recent study, (R)- and (S)-thalidomide 5% glucose intravenous solutions were successfully prepared and used in healthy male volunteers. Unfortunately the bioavailability (F) of oral formulations was not determined.[50] Absolute bioavailability has been estimated to be 93 and 67% for rats and rabbits respectively.[17] It was 64% in cynomolgus monkeys (Teo S, unpublished data). The intravenous study showed clearance and elimination half-life (t1/2 ) which were similar to those after oral administration, suggesting that F was close to 100% for the (R)-enantiomer and 50% for the (S)-enantiomer.[50] 6.3 Single-Dose Studies in Healthy and Patient Populations

Single oral dose studies with thalidomide 100–800mg have been performed in healthy and patient populations using various tablet and capsule formulations.[17,23-27] Pharmacokinetic parameters were similar between healthy and patient populations (table I and table II). The capsule formulation used in the healthy population is different from the commercial Thalomid® capsule formulation taken by HIV-seropositive patients.[19,23,25] Pharmacokinetic parameters were generally similar in healthy, prostate cancer and HIV-seropositive populations given 200mg, although two of the studies did not stabilise their plasma samples with buffer (table I). Absorption was slow, with peak concentration (Cmax) reached by 3–4 hours (tmax). Cmax, AUC to infinity (AUCβ ), terminal half-life (t1/2 ), apparent (‘oral’) clearance (CL/F) and calculated apparent volume of distribution (V/F) were 2 mg/L, 20 mg ± h/L, 5–9 hours, 7–10 L/h and 70–120L respectively. The apparently longer t1/2 of 8.70 hours from the healthy volunteer study could be due to it being a different formulation, the plasma samples not stabilised with buffer or too few samples taken during the terminal elimination phase.[17] Short (1.61 τ 1.62 hours) and long (5.2 τ 1.6 hours) half-lives representing the absorption and elimination phases © 2004 Adis Data Information BV. All rights reserved.

were reported in the HIV-seropositive study, but this representation is most probably erroneous and would result in an incorrect overestimate of V/F, as discussed below.[25] In more recent studies using Thalomid®, plasma profiles of healthy male and female volunteers taking thalidomide 200mg showed one-compartment pharmacokinetics with first-order absorption and elimination (table II).[19,20] Compartmental analysis of individual profiles were fitted using the modified equation 1: ⎛ ⎞ dose • k fast Cpt = ⎜⎜ ⎟• / V • k − k F ( fast slow) ⎟⎠ ⎝ ⎛ − kslow ⎜e ⎝

(

• t − t lag

) − e− k fast • ( t − t lag) ⎞ ⎟

⎠ (Eq. 1) where Cpt is the plasma concentration of thalidomide at time t, kfast is the faster rate constant, kslow is the slower rate constant and tlag is the time before absorption begins. Typically, the faster rate constant is assigned to the absorption term (ka) and the slower rate constant to the elimination term (kel). We substituted kfast and kslow for ka and kel to reflect this. Absorption rate-limited elimination is characteristic of the ‘flip-flop’ phenomenon, where the elimination rate is faster than the absorption rate. The flip-flop could be due to the low aqueous solubility of thalidomide and its resulting poor dissolution in the gastrointestinal tract. The poor solubility and dissolution are also demonstrated by the observed tlag of 20–40 minutes. Traditional pharmacokinetic analysis assumes that absorption is the faster process unless intravenous or oral solution data suggest otherwise, or t1/2 increases without increasing AUC values. In a flip-flop situation, however, the faster rate constant represents the elimination phase.[19] The long half-life of 5.34 τ 1.90 hours calculated from t1/2,slow = 0.693/kslow is therefore due to slower absorption. Alternatively, it is possible that the faster rate constant reflects distribution and that the terminal rate constant reflects a hybrid of both absorption and elimination until absorption becomes so slow that it truly becomes rate-limiting at 5–8 hours or longer. To study the impact of the flip-flop phenomenon, additional thalidomide profiles were simulated Clin Pharmacokinet 2004; 43 (5)

© 2004 Adis Data Information BV. All rights reserved.

3.4 τ 1.8

Capsule

No

200

1.15 τ 0.20

4.39 τ 1.27

Formulation

Plasma stabiliseda

Dose (mg)

Cmax (mg/L)

tmax (h)

0.60 τ 0.22

Ae (%)c

Calculated as Ae/AUCβ .

Calculated as CL/F divided by ; for ‘true’ V/F value, see section 6.3.

c

d

66.9 τ 34.3

7.41 τ 2.05

6.52 τ 3.81

0.13 τ 0.07

166 τ 84

7.21 τ 2.89

18.3 τ 14.1

0.06 τ 0.04

74.6 τ 16

10.7 τ 2.0

5.2 τ 1.6

1.61 τ 1.62

0.23 τ 0.32

20.1 τ 2.9

146 τ 93

13.83 τ 7.79

8.28 τ 6.00

0.12 τ 0.06

4.71b

4.09b

1.95 τ 0.38 4.2 τ 0.84

800

No

Unknown (probably tablet)

34

28–72

Both

Glioma[27]

200

Yes

Thalomid® capsule

16

24–41

Male

HIV-seropositive[25]

Ae = amount of unchanged thalidomide excreted into the urine over time; AUCβ = area under the plasma concentration-time curve from time 0 to infinity; CL/F = apparent clearance; CLR = renal clearance; Cmax = maximum plasma concentration; n = number of subjects; tlag = absorption lag time; tmax = time to Cmax; t1/2abs = absorption half-life; t1/2 = terminal elimination half-life; V/F = apparent volume of distribution; = elimination rate constant.

Median.

Plasma was stabilised by Sorensen’s citrate buffer pH 1.5.[30]

b

a

V/F (L)

87.8 τ 12.9

0.08 τ 0.03

CLR (L/h)

121 τ 5

9.2 τ 1.2

10.41 τ 2.04

CL/F (L/h)

d

6.5 τ 3.4

8.70 τ 4.11

0.34 τ 0.11

0.45 τ 0.25

11.05 τ 1.51

4.40b

4.42b

3.32b

800

1.97b

11

200

Yes

Unknown (probably tablet)

Thalomid® capsule

Yes

13

55–80

Male

Prostate cancer[24]

8 male, 1 female

26–50

t1/2 (h)

t1/2abs (h)

tlag (h)

(h–1 )

0.41 τ 0.17

1.17 τ 0.21

8

n

AUCβ (mg ± h/L)

100

21–43

Age (years)

Both

Male

Sex

HIV-seropositive[23]

Healthy[17]

Parameter

Table I. Single-dose pharmacokinetics of thalidomide in various human populations. Values are means τ SD, except where indicated

Thalidomide 317

Clin Pharmacokinet 2004; 43 (5)

318

Teo et al.

Table II. Pharmacokinetic parameters after a single dose of thalidomide 200mg (as Thalomid® capsules) in healthy male volunteers aged 20–43 years (n = 16). Plasma samples were stabilised with buffer. Values are means τ SD[19] (reproduced with permission of Sage Publications) Parameter

Noncompartmental

Cmax (mg/L)

2.00 τ 0.55

tmax (h)

3.2 τ 1.4

AUCt (mg ± h/L)

18.10 τ 3.09

AUCβ (mg ± h/L)

19.80 τ 3.61

(h–1)

Compartmentala

0.125 τ 0.036

t1/2 (h)

6.17 τ 2.56

CL/F (L/h)

10.50 τ 2.10

10.9 τ 1.7

V/F (L)b

89.3 τ 25.7

16.0 τ 11.8

kfast (h–1)

2.14 τ 3.52

t1/2fast (h)

1.03 τ 0.79

tlag (h)

0.337 τ 0.311

kslow (h–1)

0.143 τ 0.041

t1/2slow (h) a n = 17.

5.34 τ 1.90

10

IV bolus IV infusion Fast absorption Medium fast absorption Medium slow absorption Slow absorption

a

8 6

For discussion of volume difference, see section 6.3.

AUCt = area under the plasma concentration-time curve from time 0 to t; AUCβ = area under the plasma concentration-time curve from time 0 to infinity; CL/F = apparent clearance; Cmax = maximum plasma concentration; kfast = faster rate constant; kslow = slower rate constant; tlag = absorption lag time; tmax = time to Cmax; t1/2fast = fast half-life; t1/2slow = slow half-life; t1/2 = terminal elimination half-life; V/F = apparent volume of distribution; = elimination rate constant.

wherein identical doses were given as an intravenous bolus, 1-hour intravenous infusion and as four oral doses with bioavailability of 100% and different absorption rates (figure 3). AUC values were identical for the six input functions, but Cmax occurred later with slower oral absorption rates. The terminal phase, which reflects the true elimination rate process for the two intravenous input functions as well as the fastest oral absorption rate, is replaced by the absorption rate process for the three slowest absorption rate functions because of conversion to absorption rate-limited elimination. This flip-flop phenomenon can be problematic for pharmacokineticists and clinical pharmacologists if they are not aware of the changes that are taking place. V/F values can be overestimated, because CL/F values calculated as dose F/AUC are true values. Therefore, V/F calculated as CL/F divided by ka, rather than CL/F divided by kel, overestimates V/F. The magnitude of the error is the ratio of kel/ka, and the error can cause the investigator to © 2004 Adis Data Information BV. All rights reserved.

4 Plasma concentration (mg/L)

b

assume that V/F is greater than its true value or that F is less than its true value. A comparison of oral and intravenous thalidomide pharmacokinetic studies indicates that the clearance and volume values are similar (approximately 10–12 L/h and 20L) for both routes when V/F is calculated for fast-absorption doses and when V/F is calculated correctly for doses that are absorbed more slowly.[17,19,20,28] Using the correct terminal elimination rate constant kfast, the ‘true V/F’ calculated as CL/F divided by kfast was 16.0 τ 11.8L compared with 70–170L reported in previous capsule studies (table I and table

2 0

10

b

1

0.1

0 0

5

10

15

20

25

Time (h) Fig. 3. Linear (a) and log-linear (b) simulated profiles of thalidomide 200mg after intravenous (IV) bolus, 1-hour IV infusion and as four oral doses with bioavailability of 100% and different absorption rates. The following set parameters were used in the simulations: (i) for IV bolus and infusion, kfast = 0.23 h–1, V = 20L; (ii) for oral doses, kfast, tlag and V/F were set at 0.23 h–1, 0.35h and 20L, respectively, with kslow values of 0.639 (fast absorption), 0.223 (medium fast absorption), 0.116 (medium slow absorption) and 0.077 (slow absorption) h–1. kfast = faster rate constant; kslow = slower rate constant; tlag = absorption lag time; V = volume of distribution; V/F = apparent volume of distribution.

Clin Pharmacokinet 2004; 43 (5)

Thalidomide

6.4 Multiple-Dose Studies in Healthy and Patient Populations

Multiple oral dose studies in healthy women showed that thalidomide capsules 200 mg/day over 18–21 days did not produce any accumulation.[18,20] In one study, AUC at days 1 and 21 were equivalent, indicating no change in pharmacokinetics and clearance over 3 weeks of daily administration.[18] Therefore, there are no differences between the singleand multiple-dose pharmacokinetics of thalidomide at 200 mg/day. It is not known if accumulation occurs with higher daily dosages, although pharmacokinetic simulations of 400 and 800mg once daily suggest © 2004 Adis Data Information BV. All rights reserved.

400mg, day 1 400mg, day 21 800mg, day 1 800mg, day 21

a 8

6

4

Plasma concentration (mg/L)

II).[17,19,23,25] This value is similar to the distribution volume calculated from the intravenous study.[50] No differences in the single-dose pharmacokinetics were seen between the 200mg Thalomid® capsule and tablet studies in both healthy and patient populations. There were also no differences in the pharmacokinetics between prostate and glioma patients administered thalidomide 800mg as tablets (table I). In the intravenous formulation study, a fourcompartment model representing both enantiomers was used to describe a 1-hour 50mg intravenous infusion in healthy male volunteers.[50] The model incorporated chiral inversion between the two enantiomers. Volume of distribution, clearance and t1/2 for the (R)- and (S)-enantiomers were 18 τ 7.5 and 24 τ 11L, 10 τ 2.1 and 21 τ 4.6 L/h and 4.7 τ 0.5 and 4.7 τ 0.5 hours respectively. The identical t1/2 for the (R)- and (S)-enantiomers is because each ratelimits the other due to interconversion. The volume of distribution after intravenous administration was similar to our ‘true’ oral value calculated above, indicating absorption rate-limited pharmacokinetics. t1/2 was similar between the enantiomers and also similar to those from racemic oral studies (table II). Clearance was significantly different between the two enantiomers, with the (S)-enantiomer eliminated twice as fast as the (R)-enantiomer. This is consistent with the previous finding that the (S)-enantiomer is eliminated to a greater extent than the (R)enantiomer.[35] The racemic CL/F of 10.50 L/h therefore reflects the (R)-enantiomer.

319

2

0 b 10

1

0 0

5

10

15

20

25

Time (h) Fig. 4. Linear (a) and log-linear (b) simulated oral profiles of thalidomide 400 and 800 mg/day over 21 days. Both simulations had kfast, tlag and V/F set at 0.52 h–1, 0.35h and 20L, respectively, with kslow of 0.14 and 0.105 h–1 for the 400 and 800mg doses. kfast = faster rate constant; kslow = slower rate constant; tlag = absorption lag time; V/F = apparent volume of distribution.

that this would not occur (figure 4). Steady state was achieved within 2–3 days. Although AUC values do not increase or decrease with increasing dose, absorption is delayed in this dose range, resulting in a delay in Cmax and a flip-flop phenomenon that results in greater plasma accumulation with the 800mg dose. This flip-flop phenomenon appears to be due to more protracted absorption with increasing thalidomide dose and with formulations that are not as readily available for absorption, even at lower doses.[19,21] A maximum steady-state plasma concentration (Cssmax) of 1.2 mg/L was predicted at a dosage of 200 mg/day.[28] Computer simulations of 400 and 800 mg/day over 21 days predicted Cssmax values of 3.5 and 6.0 mg/L, respectively (figure 4). Simulation Clin Pharmacokinet 2004; 43 (5)

320

Teo et al.

Table III. Pharmacokinetic parameters of thalidomide after multiple doses in patients with prostate cancer, glioma and breast cancer. Plasma samples were not stabilised with buffer. Values are means τ SD Glioma[27]

Breast[26]

Parameters

Prostate[24] 200 mg/day

200–1200 mg/day

800–1200 mg/day

200 mg/day

Formulation

Unknowna

Unknowna

Unknowna

Tablet

Tablet

Patients (n)

10

3–11

31

14

14

Age (years)

55–80

28–72

30–85

Mean duration (days)

67

Dose escalated 200 mg/day every 2 weeks to 1200 mg/day

Dose escalated 200 mg/day every 2 weeks to 1200 mg/day

2 weeks

0.10 τ 0.03

0.066 τ 0.054

0.16 τ 0.17

7.08 τ 1.87

16.19 τ 9.57

8.31 τ 7.12

(h–1)

0.39 τ 0.41

ka (h–1) t1/2 (h)

800 mg/day

CL/F (L/h)

6.35 τ 1.64

7.74 τ 2.27

12.65 τ 6.63

V/F (L)b

64.53 τ 23.20

168 τ 82

124 τ 73

a

Probably tablet.

b

For discussion of voume difference, see section 6.3.

5.4 τ 2.4

7.7 τ 3.5

CL/F = apparent clearance; ka = absorption rate constant; t1/2 = terminal elimination half-life; V/F = apparent volume of distribution (see table I); = elimination rate constant.

of the 800mg dose used three fourths the absorption rate constant (ka) for the 400mg dose to account for the protracted and slower absorption due to flipflop. Simulations by others have confirmed the Css[28] In addition, observed max for the 400mg dose. ss C max values from patients with prostate cancer, glioma and breast cancer were similar to our predictions for both dosages (table III). The pharmacokinetics of thalidomide are similar between healthy and patient populations in limited studies using different formulations. Multiple-dose and dose-escalation pharmacokinetic data are available from studies in patients with prostate and breast cancer, and prostate cancer and glioma, respectively (table III). Elderly prostate cancer patients taking thalidomide 200 mg/day as tablets over a mean of 67 days and healthy men taking a single 200mg dose of Thalomid® capsules had similar t1/2 and CL/F (table II and table III),[19,24] although Cmax and AUC were not reported for the prostate patients. Dose-escalation studies in prostate and glioma patients, starting from 200 or 800 mg/day and rising by 200 mg/day every 2 weeks to a maximum of 1200 mg/day, showed significant absorption rate-limited elimination, as demonstrated by the prolonged t1/2 . CL/F and V/F for the three cancer studies were comparable to those from healthy male volunteers taking Thalomid® 200mg capsules, indicating no diseaseinduced change in capacity to remove thalidomide © 2004 Adis Data Information BV. All rights reserved.

from the blood (table II and table III). CL/F was lower in healthy women, probably due to differences in formulations, but was similar in the prostate and breast cancer studies (prostate cancer, 7.41 τ 2.05 and 7.21 τ 2.89 L/h for 200 and 800 mg/day; breast cancer, 5.4 τ 2.4 and 7.7 τ 3.5 L/h for 200 and 800 mg/day; table I).[24,26] Clearance is therefore independent of dose after single and multiple doses in these populations (table I, table II and table III). V/F in cancer patients was similar to that in healthy volunteers taking a single 200mg dose (table II and table III). Estimates of steady-state concentrations during once-daily administration in the cancer studies were comparable and proportional to increasing dose (table IV). Cssmax for the cancer patients and simulated Cssmax for healthy women at 400 and 800 mg/ day were similar at 3.5 mg/L and 5–8 mg/L, respectively (figure 4, table IV). A higher Cssmax of 3.6 mg/L was estimated for healthy male volunteers taking a 200mg capsule every 6 hours instead of once daily, suggesting accumulation with this regimen,[17] and indeed this regimen is not recommended due to persistent adverse events (see section 6.8). Plasma concentration-time profiles from the prostate cancer and glioma studies were similar to those in the healthy population, best fitting a onecompartment linear model.[19,24,27] Clin Pharmacokinet 2004; 43 (5)

Thalidomide

321

6.5 Dose Proportionality

Dose proportionality was observed when comparing the 100mg (HIV patients) and 200mg (prostate cancer and HIV patients) doses as measured by Cmax and AUC. It was not observed for Cmax when compared with the healthy volunteer study (table I). This could be due to the differences discussed earlier (section 6.5). In a single-dose study with Thalomid® in healthy subjects, there was a less than proportional increase in Cmax and prolongation of tmax with increasing doses of 50 to 400mg (table V).[22] This is because of slow gastric absorption caused by the low aqueous solubility and dissolution characteristics of thalidomide. Once the fixed volume of gastric fluid at the absorption site has been saturated with drug, further dissolution may not take place until some of the drug is absorbed across the intestinal lumen. This could explain the increased tmax and decreased Cmax with increasing dose. In contrast, thalidomide exhibits dose-proportional increase in AUC, suggesting that the amount absorbed and its clearance are independent of the dose used (figure 5, table V).[22] The increase in V/F between 200 and 400mg is probably due to it being calculated using ka rather than the ‘true’ kel. Taking into account flip-flop, the ‘true V/F’ was previously estimated to be 16L. The decrease in the terminal rate constant ( ) suggests that the absorption of thalidomide is dependent on its solubility in gastric fluid. It is likely that the trends observed in this study would also be present at doses higher than 400mg. Specifically, in a linear system, as doses increase from 50 to

200 to 400mg, Cmax and AUC would be expected to increase by factors of 4 and 8, respectively, and tmax would be unchanged. In reality, mean Cmax only increased 2.8-fold and 4.6-fold, whereas AUCβ increased 3.9-fold and 7.4-fold; tmax increased from 2.9 to 3.5 to 4.3 hours (table V). Based on these trends, Cmax, AUC and tmax values for an 800mg dose would be expected to be approximately 3.8 mg/L, 65 mg ± h/L and 5.1 hours, respectively.[22] 6.6 Effects of Age, Sex, Smoking and Food

A recent report in patients with gliomas suggested an inverse relationship between age and clearance.[27] This finding has not been verified. Sex and smoking do not appear to affect the single- and multiple-dose pharmacokinetics of thalidomide, since pharmacokinetic profiles and parameters were similar among these populations (table VI).[19,21,22,25] A high-fat meal resulted in a tlag of 0.5–1.5 hour, a 62% delay in mean tmax, an 8.54% increase in Cmax and a 5.5% decrease in AUC (table VI, figure 6).[21] Fatty foods generally delay gastric emptying, thereby providing ample time for greater dissolution and absorption. Food therefore has little effect on the extent of absorption of thalidomide capsules. 6.7 Elimination

No comprehensive elimination studies have been performed in humans. Urinary excretion of parent thalidomide was found to be less than 1% of the total dose over 24 hours, with renal clearance estimated to be 0.08 τ 0.03 L/h (table I).[17] No products of

Table IV. Maximum steady-state plasma concentration of thalidomide after multiple doses in patients with prostate cancer, glioma and breast cancer. Plasma samples were not stabilised with buffer. Values are means τ SD Parameter

Maximum steady-state plasma concentration (mg/L) prostate[24]

glioma[27]

breast[26]

Formulation

Unknown (probably tablet)

Unknown (probably tablet)

Tablet

Number of patients

3–11

31

14

Age (years)

55–80

28–72

30–85

Dose (mg/day) 200

1.81 τ 0.81

400

3.43 τ 0.79

600

5.58 τ 1.77

1.52 τ 1.10

800

7.57 τ 1.83

4.97 τ 1.96

1000

9.18 τ 1.95

5.95 τ 2.57

1200

11.07 τ 2.47

7.40 τ 3.21

Not measured

2 partial, 2 minor, 12 stable disease

Tumour response

© 2004 Adis Data Information BV. All rights reserved.

6.2 τ 4.3

None

Clin Pharmacokinet 2004; 43 (5)

322

Teo et al.

Table V. Pharmacokinetic parameters of thalidomide after oral doses of 50, 200 and 400mg (as Thalomid® capsules) in healthy subjects 20–54 years of age (n = 15). Plasma samples were stabilised with buffer. Values are means τ SD[22] (reproduced with permission of Sage Publications) Parameter

Dose (mg) 800a

50

200

400

Cmax (mg/L)

0.62 τ 0.32

1.76 τ 0.52

2.82 τ 0.80

3.8

tmax (h)

2.9 τ 1.9

3.5 τ 2.0

4.3 τ 1.6

5.1

AUCt (mg ± h/L)

3.44 τ 0.66

15.6 τ 3.8

31.4 τ 9.7

AUCβ (mg ± h/L)

4.90 τ 0.77

18.9 τ 3.3

36.4 τ 9.6

AUC extrapolated (%)

30%

17%

14% 0.104 τ 0.030

0.138 τ 0.040

0.132 τ 0.032

t1/2 (h)

5.5 τ 2.0

5.5 τ 1.4

7.3 τ 2.6

CL/F (L/h)

10.4 τ 1.8

10.9 τ 1.9

11.7 τ 2.8

87.8 τ 20.2

121.9 τ 47.5

(h–1)

a

81.1 τ 26.2 Approximated values, see section 6.5.

b

For discussion of volume difference, see section 6.3.

V/F (L)b

65

AUCt = area under the plasma concentration-time curve from time 0 to t; AUCβ = area under the plasma concentration-time curve from time 0 to infinity; CL/F = apparent clearance; Cmax = maximum plasma concentration; tmax = time to Cmax; t1/2 = terminal elimination half-life; V/F = apparent volume of distribution; = elimination rate constant.

enzymatic hydrolysis were detected in either the plasma or urine. As discussed in section 5, numerous products of spontaneous hydrolysis are excreted in the urine. Most of an absorbed oral dose in humans is eliminated in the excreta, but it is not known how much is eliminated through urine and faeces. Studies in rats, rabbits and monkeys using radiolabelled thalidomide showed that most elimination was passive and through the urine. After a single oral dose of 10 mg/kg, urine and faecal eliminations over 48 hours were 8 and 14% for rats, 87 and 11% for rabbits and 94 and 2% for monkeys.[47,48] 6.8 Adverse Events

Adverse events seen in healthy populations taking single doses of Thalomid® capsules included sedation, dizziness and headache.[19,21,22] No serious or unexpected adverse events were evident. Healthy patients taking multiple doses of thalidomide 200 mg/day for 3 weeks exhibited mild to moderate sedation, headache, dizziness, constipation, paraesthesia and dry mouth.[18,20] Prostate cancer patients taking 200–1200 mg/day had peripheral neuropathy, sedation, pulmonary embolism and shortness of breath.[24] Recently, some newly diagnosed multiple myeloma patients taking thalidomide 400 mg/day along with certain chemotherapeutic agents exhibited deep vein thrombosis. The aetiology of thrombo© 2004 Adis Data Information BV. All rights reserved.

embolic events in cancer patients could be due to tumour cell interactions with pre-existing coagulation factors, generation of procoagulant factors and activation of platelet and/or vascular endothelium.[51] The significance of thalomide in the thrombosis is unknown as these patients were also taking other medications. The thrombosis was successfully treated with anticoagulation therapy (see section 8.2).[52] Thalidomide could produce adverse events and/ or toxicity through exposure in bodily fluids other than blood. We recently reported on the presence of thalidomide in human semen after oral doses of 100 mg/day for 8 weeks in HIV patients.[53] Plasma and semen levels of 10–350 g/L and 10–250 g/kg were obtained after 4 and 8 weeks of administration. Semen levels could be significantly higher at greater therapeutic dosages. We have also documented the presence of thalidomide in rabbit milk after oral administration (data not shown). Thalidomide is therefore distributed into major bodily fluids. 7. Concentration-Response Relationship In a multiple-dose study assessing the effects of thalidomide 200 mg/day at night for 18 days on the pharmacokinetics of oral contraceptives in healthy premenopausal females, headache, asthenia, dizziness, constipation, dry mouth and paraesthesia were seen in 82, 55, 45, 36, 36 and 36% of the volunteers, Clin Pharmacokinet 2004; 43 (5)

Thalidomide

Potential drug-drug interactions with thalidomide have not been extensively studied. Only two controlled interaction studies have been performed, both with oral contraceptives, and no pharmacokinetic interactions were observed. Since thalidomide is spontaneously hydrolysed at physiological pH, interactions mediated through the hepatic CYP system are not likely to be significant. Thalidomide could, however, potentially interact with coadministered medications by other mechanisms. © 2004 Adis Data Information BV. All rights reserved.

Plasma concentration (mg/L)

As part of Celgene’s mandatory safety program (System for Thalidomide Education and Prescribing Safety [STEPS®]), female patients of childbearing age are required to use two forms of contraception.[54] The effectiveness of hormonal oral contraceptives has previously been shown to be diminished by drugs that induce hormone metabolism.[55-57] Oral contraceptive steroids are mainly broken down by hepatic CYP3A4. Coadministration of drugs that inhibit this isozyme can alter hormone metabolism.[56,58] Two recent studies in healthy women showed that thalidomide does not alter the pharmacokinetics of ethinylestradiol and norethindrone.[18,20] The first was an open-label crossover study where the pharmacokinetics of single doses of ethinylestradiol 0.07mg and norethindrone 2mg were measured at baseline and after 3 weeks of thalidomide capsules 200 mg/day. No changes in the pharmacokinetics of 10

50 mg 200 mg 800 mg

a

1

0.1

0.01 0

10

20

30

40

Time (h)

10

b

Cmax AUC

6

3

5

Cmax (mg/L)

8. Pharmacokinetic Interactions

8.1 Oral Contraceptives

AUC0-β (mg ± h/L)

respectively. No accumulation was seen and a mean plasma concentration of 0.94 τ 0.19 mg/L was obtained at day 18.[20] A correlation analysis between plasma concentrations of the favoured (R)-enantiomer (AUC(R)/AUC(S) = 1.6) and sedation was performed to estimate efficacy concentrations.[33,35] Although no temporal relationship was reported, sleep was marginally (p = 0.057) correlated with blood (R)-thalidomide concentration. The predicted concentration associated with a 20% probability of sleep was 0.77 mg/L (95% CI 0.41–3.96 mg/L).[50] Blood concentration associated with a 50% probability of ‘moderate’ tiredness was 1.6 mg/L. There are currently no controlled studies that correlate blood or plasma thalidomide concentrations with efficacy. Two uncontrolled multiple-dose studies in breast cancer and glioma patients have attempted to correlate thalidomide concentration with tumour response.[26,27] Two partial responses, two minor responses and 12 stable diseases were reported at up to 74 weeks from a group of 31 glioma patients treated with thalidomide 800 mg/ day increasing by 200 mg/day every 2 weeks to a final maintenance dosage of 1200 mg/day. The study, however, did not specify the dosage level at which these tumour responses were seen. Steadystate plasma concentrations ranged from 5 to 7 mg/L at 800–1200 mg/day (table IV).[27] Adverse events similar to those in the contraceptives study[20] were evident at these higher plasma concentrations, although the incidences were not clear. In the other study, patients with metastatic breast cancer did not show any tumour response at 800 mg/day with a steady-state concentration of 6.2 τ 4.3 mg/L after 8 weeks of treatment.[26]

323

0

0 0

100

200

300

400

500

Dose (mg) Fig. 5. (a) Log-linear plots of plasma profiles for three doses of thalidomide and (b) dose-normalised area under the concentrationtime curve (AUC) and maximum concentration (Cmax) versus dose for 15 healthy males and females (adapted from Teo et al.,[22] with permission of Sage Publications).

Clin Pharmacokinet 2004; 43 (5)

3.21 τ 0.07 3.05 τ 0.25 3.24 τ 0.09 95

3.11 τ 0.35 89–102 3.15 τ 0.18 3.19 τ 0.22 ln AUCβ

95

3.05 τ 0.36 88–103 3.07 τ 0.25 3.11 τ 0.23

108

© 2004 Adis Data Information BV. All rights reserved.

AUCt = area under the plasma concentration-time curve from time 0 to t; AUCβ = area under the plasma concentration-time curve from time 0 to infinity; Cmax = maximum plasma concentration; MRT = mean residence time; tmax = time to Cmax; t1/2 = terminal elimination half-life; = elimination rate constant.

0.841 τ 0.244

3.16 τ 0.08 2.93 τ 0.37

11.0 τ 2.6

ln AUCt

3.15 τ 0.11

0.603 τ 0.148

10.8 τ 0.8

0.573 τ 0.215 0.75 τ 0.24 0.66 τ 0.21

0.725 τ 0.194

10.9 τ 2.3

96–122 10.9 τ 2.0 10.4 τ 2.3 MRT (h)

ln Cmax

10.1 τ 2.5

5.00 τ 1.21

0.145 τ 0.030 0.135 τ 0.020

5.22 τ 0.81 6.09 τ 2.09

0.122 τ 0.026

5.35 τ 0.86

0.132 τ 0.019

0.141 τ 0.025

5.09 τ 1.03

0.126 τ 0.023 t1/2 (h)

5.80 τ 1.72

–5.47 23.5 τ 3.7 24.7 τ 5.1 AUCβ (mg ± h/L)

(h–1)

23.6 τ 2.0

24.9 τ 1.7 21.7 τ 5.1 25.5 τ 2.3

6.25 τ 2.55

–4.83 22.1 τ 4.3 23.0 τ 5.1

23.4 τ 8.1

19.6 τ 6.0 23.6 τ 2.6

2.38 τ 0.53

AUCt (mg ± h/L)

22.2 τ 8.1

5.80 τ 2.17

Fed

1.84 τ 0.25 2.10 τ 0.41

3.75 τ 1.28 4.50 τ 0.58 61.8 6.08 τ 2.33 4.00 τ 1.13

1.81 τ 0.38 8.54 2.17 τ 0.51 1.99 τ 0.41 Cmax (mg/L)

tmax (h)

males difference between means (%) 90% CI of ratio (both sexes) (both sexes)

mean ratio

Comparison of fed versus fasted Fed Fasted

Fasted

females

males

females

Teo et al.

Parameter

Table VI. Comparative pharmacokinetics of thalidomide between fed and fasted conditions in healthy volunteers aged 20–46 years after an oral 200mg dose of thalidomide (as Thalomid® capsules) [n = 13; 5 males and 8 females]. Plasma samples were stabilised with buffer. Values are means τ SD [21] (reproduced with permission of John Wiley & Sons Limited)

324

ethinylestradiol and norethindrone were seen. Mean (τ SD) AUCβ for ethinylestradiol at baseline and after thalidomide dosing were 6580 τ 1100 and 5970 τ 1560 ng ± h/L, respectively. Corresponding norethindrone values were 103 τ 54 and 107 τ 58 g ± h/L. CL/F and t1/2elim also did not change for both drugs.[20] The second study was also a cross-over study and looked at the effects of steady-state plasma thalidomide concentrations on the pharmacokinetics of the two hormones.[18] Subjects were first given thalidomide 200 mg/day as capsules for 21 days, followed by a single dose of ethinylestradiol 35 g and norethindrone 1mg. They were then given a single dose of the hormones without prior administration of thalidomide. Coadministration of thalidomide at steady state did not affect the pharmacokinetics of either oral contraceptive hormone. These two studies support our previous finding that thalidomide is minimally metabolised by hepatic CYP.[46] Thalidomide will therefore not alter the plasma concentrations or clinical efficacy of these contraceptives. These two studies however only looked at single doses of the hormones. The effects of thalidomide on the multiple-dose pharmacokinetics of ethinylestradiol and norethindrone are unknown. In particular, women who are taking thalidomide and who also require treatment with rifampicin (rifampin), rifabutin, barbiturates, glucocorticoids, phenytoin, carbamazepine or protease inhibitors should not rely only on hormonal contraception, since these agents have been shown to induce the metabolism of oral contraceptives, thereby reducing their efficacy.[59] 8.2 Warfarin

Repeated cycles of combination chemotherapy with thalidomide, glucocorticoid (dexamethasone) and cytotoxic drugs (vincristine, doxorubicin, cyclophosphamide, etoposide, cisplatin) for newly diagnosed multiple myeloma have been shown to increase the risk of deep-vein thrombosis, which can be controlled by heparin and warfarin therapy.[52] This increase was not seen when thalidomide was used as a single agent.[60] Warfarin is a commonly used oral anticoagulant, has a long half-life of 40–70 hours and is metabolClin Pharmacokinet 2004; 43 (5)

Thalidomide

8.3 Other Drugs

A recent study in glioma patients taking thalidomide along with rifampicin, phenobarbital, cimetidine or diltiazem showed no effect on thalidomide clearance.[27] These concomitant drugs are inducers of hepatic CYP enzymes, which do not play a role in thalidomide breakdown. No changes in oral availability are therefore expected when taking thalidomide with grapefruit juice, a known inhibitor of the major CYP3A4 isoform expressed in adult human small intestine and liver.[62,63] Thalidomide has been used for nearly 40 years in various oncologic and inflammatory conditions, either alone or in conjunction with a variety of drugs. It is currently being used experimentally in over 150 clinical trials in combination with other chemotherapeutics (Teo S, unpublished data). Few pharmacokinetic interactions have been reported. There have been no reports of increased dosage requirements or toxicity when used along with dapsone and rifampin for ENL. Early studies in mice have shown that thalidomide potentiates the actions of other CNS depressants such as the barbiturates. Thalidomide should therefore not be used along with other hypnosedatives. Thalidomide also increases the oral toxicity of ethanol in mice and the duration of catatonia produced by chlorpromazine and reserpine. CNS stimulants such as metamfetamine and methylphenidate have been shown to counteract the depressant action of thalidomide.[2] © 2004 Adis Data Information BV. All rights reserved.

An adverse effect of prolonged thalidomide use is the development of sensory peripheral neuropathy.[64] The neuropathy could be worsened with concomitant use of chemotherapeutics such as docetaxel and cisplatin that also cause neuropathy. 9. Hepatic and Renal Dysfunction No comprehensive studies have been performed to assess the pharmacokinetics of thalidomide in patients with hepatic dysfunction. Since thalidomide is spontaneously and nonspecifically hydrolysed, plasma concentrations should not change in patients with compromised liver function. In animal studies, >80% of an oral thalidomide dose is passively eliminated through the kidneys as hydrolysis products (see section 6.7). Hence, the pharmacokinetics of thalidomide should also not change in renally impaired patients. In a study of thalidomide in patients with impaired renal function or during dialysis, the pharmacokinetics of thalidomide in patients with renal failure were similar to previously reported values in patients with normal renal function. Clearance during dialysis was doubled. The investigators concluded that the thalidomide dosage need not be changed for patients with decreased kidney function, and that there is no need for a supplementary dose after haemodialysis.[65] 10. Conclusions Even though it has been over 40 years since thalidomide was initially prescribed, it is only reFasted Fed

10 Plasma concentration (mg/L)

ised by the liver. It is highly bound (99%) with great affinity to one (site I) of the two binding sites of human serum albumin.[61] Thalidomide is not as highly bound (55–65%) to human serum albumin, and the binding site and affinity are not known.[34] There is no information on the potential interaction between thalidomide and warfarin. The current practice of stopping thalidomide administration should any symptoms of hypercoagulability occur, initiating anticoagulation therapy and then resuming treatment once symptoms have subsided has allowed patients to continue receiving thalidomide. Prophylactic anticoagulation has been recommended for patients taking thalidomide in combination with chemotherapy.[52]

325

1

0.1

0.01

0.001 0

10

20

30

40

Time (h) Fig. 6. Log-linear plasma profiles for thalidomide after oral administration of 200mg to 13 healthy males and females under fed and fasted conditions (adapted from Teo et al.,[21] with permission of John Wiley & Sons Limited).

Clin Pharmacokinet 2004; 43 (5)

326

Teo et al.

cently that its clinical pharmacokinetics have been elucidated. Thalidomide exhibits one-compartment pharmacokinetics with first-order absorption and elimination. There is little difference in the multiple dose pharmacokinetics of healthy and patient populations between doses of 200–800 mg/day. The low solubility of thalidomide in the gastrointestinal tract is the cause for its elimination rate being faster than its absorption rate (flip-flop). It exhibits dose proportional incease in AUC from 50–400mg which brackets most current therapeutic doses. Acknowledgements This review is dedicated to the memory of Wayne Colburn PhD who passed away on February 2, 2003. Wayne will be remembered for his leadership and passion in advancing the profession of Clinical Pharmacology and for his enormous contribution to this discipline. This review was sponsored by Celgene Corporation. Doctors Teo, Stirling, Jaworsky, Scheffler, Thomas and Laskin are employees of Celgene Corporation.

References 1. Koch H. Thalidomide and congeners as anti-inflammatory agents. Prog Med Chem 1995; 22: 165-242 2. Somers GF. Pharmacological properties of thalidomide (• phthalidimo glutarimide), a new sedative hypnotic drug. Br J Pharmacol 1960; 15: 111-6 3. Neuhaus G, Ibe K. Survival following suicide attempt with 14 100mg of thalidomide. Dis Nerv Syst 1961; 22: 52-3 4. Mellin GW, Katzenstein M. The saga of thalidomide: neuropathy to embryopathy, with case reports and congenital abnormalities. N Engl J Med 1963; 267: 1184-93 5. Schumacher H, Blake D, Gurian J, et al. A comparison of the teratogenic activity of thalidomide in rabbits and rats. J Pharmacol Exp Ther 1968; 160: 189-99 6. Szabo K, Steelman R. Effects of maternal thalidomide treatment on pregnancy, fetal development and mortality of the offspring in random-bred mice. Am J Vet Res 1967; 28: 1823-8 7. Kelsey F. Thalidomide update: regulatory aspects. Teratology 1988; 38: 221-6 8. Sheskin J. Thalidomide in the treatment of lepra reaction. Clin Pharmacol Ther 1965; 6: 303-6 9. Calabrese L, Fleischer A. Thalidomide: current and potential clinical applications. Am J Med 2000; 108: 487-95 10. Teo S, Resztak K, Scheffler M, et al. Thalidomide in the treatment of leprosy. Microbes Infect 2002; 4: 1193-202 11. Singhal S, Mehta J, Desikan R, et al. Antitumor activity of thalidomide in refractory multiple myeloma. N Engl J Med 1999; 341: 1565-71 12. Srinavas S, Guardino A. Randomized phase II trial of high and low dose thalidomide in metastatic renal cell carcinoma [abstract]. Proc Am Soc Clin Oncol 2002; 38: 2403 13. Marx G, Pavlakis N, McCowatt S, et al. Phase II study of thalidomide in the treatment of recurrent glioblastoma multiforme. J Neurooncol 2001; 54: 31-8 14. Govindarajan R, Heaton K, Broadwater R, et al. Effect of thalidomide on gastrointestinal toxic effects of irinotecan. Lancet 2000; 356: 566-7

© 2004 Adis Data Information BV. All rights reserved.

15. Moreira A, Sampaio E, Zmuidzinas A, et al. Thalidomide exerts its inhibitory action on TNF-• by enhancing mRNA degradation. J Exp Med 1993; 177: 1675-80 16. D’Amato RJ, Loughnan MS, Flynn E, et al. Thalidomide is an inhibitor of angiogenesis. Proc Natl Acad Sci U S A 1994; 91: 4082-5 17. Chen T-L, Vogelsang G, Petty B, et al. Plasma pharmacokinetics and urinary excretion of thalidomide after oral dosing in healthy male volunteers. Drug Metab Dispos 1989; 17: 402-5 18. Trapnell CB, Donahue SR, Collins JM, et al. Thalidomide does not alter the pharmacokinetics of ethinyl estradiol and norethindrone. Clin Pharmacol Ther 1998; 64: 597-602 19. Teo SK, Colburn WA, Thomas SD. Single-dose oral pharmacokinetics of three formulations of thalidomide in healthy male volunteers. J Clin Pharmacol 1999; 39: 1162-8 20. Scheffler M, Colburn W, Kook K, et al. Thalidomide does not alter estrogen-progesterone hormone single-dose pharmacokinetics. Clin Pharmacol Ther 2000; 65: 483-90 21. Teo S, Scheffler M, Kook K, et al. Effect of a high-fat meal on thalidomide pharmacokinetics and the relative bioavailability of oral formulations in healthy men and women. Biopharm Drug Dispos 2000; 21: 33-40 22. Teo S, Scheffler M, Kook K, et al. Thalidomide single-dose proportionality assessment in healthy subjects. J Clin Pharmacol 2001; 41: 662-7 23. Piscitelli SC, Figg WD, Hahn B, et al. Single-dose pharmacokinetics of thalidomide in human immunodeficiency virusinfected patients. Antimicrob Agents Chemother 1997; 41: 2797-9 24. Figg WD, Raje S, Bauer KS, et al. Pharmacokinetics of thalidomide in an elderly prostate cancer population. J Pharm Sci 1999; 88: 121-5 25. Noormohamed FH, Youle MS, Higgs CJ, et al. Pharmacokinetics and hemodynamic effects of single oral doses of thalidomide in asymptomatic human immunodeficiency virusinfected subjects. AIDS Res Hum Retroviruses 1999; 15: 1047-52 26. Baidas S, Winer E, Fleming G, et al. Phase II evaluation of thalidomide in patients with metastatic breast cancer. J Clin Oncol 2000; 18: 2710-7 27. Fine H, Figg W, Jaeckle K, et al. Phase II trial of the antiangiogenic agent thalidomide in patients with recurrent high-grade gliomas. J Clin Oncol 2000; 18: 708-15 28. Eriksson T, Bjorkman S, Hoglund P. Clinical pharmacology of thalidomide. Eur J Pharmacol 2001; 57: 365-76 29. Teo S, Chandula R, Harden J, et al. Sensitive and rapid method for the determination of thalidomide in human plasma and semen using solid-phase extraction and liquid chromatography-tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci 2002; 767: 145-51 30. Eriksson T, Bjorkman S, Fyge A. Determination of thalidomide in plasma and blood by high-performance liquid chromatography: avoiding hydrolytic degradation. J Chromatogr 1992; 582: 211-6 31. Kenyon B, Browne F, D’Amato R. Effects of thalidomide and related metabolites in a mouse corneal model of neovascularization. Exp Eye Res 1997; 64: 971-8 32. Reist M, Carrupt P-A, Francotte E, et al. Chiral inversion and hydrolysis of thalidomide mechanisms and catalysis by bases and serum albumin and chiral stability of teratogenic metabolites. Chem Res Toxicol 1998; 11: 1521-8 33. Hoglund P, Eriksson T, Bjorkman S. A double-blind study of the sedative effects of the thalidomide enantiomers in humans. J Pharmacokinet Biopharm 1998; 26: 363-83 34. Eriksson T, Bjorkman S, Roth B, et al. Enantiomers of thalidomide: blood distribution and the influence of serum albumin on chiral inversion and hydrolysis. Chirality 1998; 10: 223-8

Clin Pharmacokinet 2004; 43 (5)

Thalidomide

35. Eriksson T, Bjorkman S, Roth B, et al. Stereospecific determination, chiral inversion in vitro and pharmacokinetics in humans of the enantiomers of thalidomide. Chirality 1995; 7: 44-52 36. Wnendt S, Finkam M, Winter W, et al. Enantioselective inhibition of TNF-• release by thalidomide and thalidomide-analogues. Chirality 1996; 8: 390-6 37. Turk B, Jiang H, Liu J. Binding of thalidomide to • 1-acid glycoprotein may be involved in its inhibition of tumor necrosis factor • production. Proc Natl Acad Sci U S A 1996; 93: 7552-6 38. Iyer C, Languillon J, Ramanujan K, et al. WHO coordinated short-term double-blind trial with thalidomide in the treatment of acute lepra reactions in male lepromatous patients. Bull World Health Organ 1971; 45: 719-32 39. Ehrenpreis E, Kane S, Cohen L, et al. Thalidomide therapy for patients with refractory Crohn’s disease: an open label trial. Gastroenterology 1999; 117: 1271-7 40. Braun A, Harding F, Weinreb S. Teratogen metabolism of thalidomide activation is mediated by cytochrome P-450. Toxicol Appl Pharmacol 1986; 82: 175-9 41. Tsambaos D, Bolsen K, Georgiou S, et al. Effects of oral thalidomide on rat liver and skin microsomal P450 isozyme activities on urinary porphyrin excretion: interaction with oral hexachlorobenzene. Arch Dermatol Res 1994; 286: 347H-9H 42. Teo S, Trigg N, Shaw M, et al. Subchronic toxicity of thalidomide in rodents after 13 weeks of oral administration. Int J Toxicol 1999; 18: 337-52 43. Knoche B, Blaschke G. Stereoselectivity of the in vitro metabolism of thalidomide. Chirality 1994; 6: 221-4

327

51. Francis J, Biggerstaff J, Amirkhorsravi A. Hemostatis and malignancy. Semin Thromb Hemost 1998; 24: 93-109 52. Zangari M, Anaissie E, Barlogie B, et al. Increased risk of deepvein thrombosis with multiple myeloma receiving thalidomide and chemotherapy. Blood 2001; 98: 1614-5 53. Teo S, Harden J, Burke A, et al. Thalidomide is distributed into human semen after oral dosing. Drug Metab Dispos 2001; 29: 1355-7 54. Zeldis J, Williams B, Thomas S, et al. STEPS: a comprehensive program for controlling and monitoring access to thalidomide. Clin Ther 1999; 21: 319-30 55. Rosenfeld W, Doose D, Walker S, et al. Effect of topiramate on the pharmacokinetics of an oral contraceptive containing norethindrone and ethinyl estradiol in patients with epilepsy. Epilepsia 1997; 38: 317-23 56. Saano V, Glue P, Banfield C, et al. Effects of felbamate on the pharmacokinetics of a low-dose combination oral contraceptive. Clin Pharmacol Ther 1995; 58: 523-32 57. Shenfield G. Oral contraceptives: are drug interactions of clinical significance? Drug Saf 1993; 9 (1): 21-37 58. Guengerich F. Inhibition of oral contraceptives steroid-metabolizing enzymes by steroids and drugs. Am J Obstet Gynecol 1990; 163: 2159-63 59. Lacy C, Armstrong L, Ingrim N, Lance L, editors. Drug information handbook. Cleveland: Lexi-Comp Inc, 1997-98 60. Barlogie B, Zangari M, Spencer T, et al. Thalidomide in the management of multiple myeloma. Semin Hematol 2001; 38: 250-9

44. Ando Y, Fuse E, Figg W. Thalidomide metabolism by the CYP2C family. Clin Cancer Res 2002, 73

61. Kosa T, Maruyama T, Otagiri M. Species difference of serum albumins: I. drug binding sites. Pharm Res 1997; 14: 1607-13

45. Eriksson T, Bjorkman S, Roth B, et al. Hydroxylated metabolites of thalidomide: formation in vitro and in vivo in man. J Pharm Pharmacol 1998; 50: 1409-16

62. Shimada T, Yamazaki H, Mimura M, et al. Interindividual variations in human liver cytochrome P-450 enzymes involved in the oxidation of drugs, carcinogens and toxic chemicals: studies with liver microsomes of 30 Japanese and 30 Caucasians. J Pharmacol Exp Ther 1994; 270: 414-23

46. Teo S, Sabourin P, O’Brien K, et al. Metabolism of thalidomide in human microsomes, cloned human cytochrome P-450 isozymes and Hansen’s disease patients. J Biochem Mol Toxicol 2000; 14: 140-7 47. Schumacher H, Smith R, Williams R. The metabolism of thalidomide: the fate of thalidomide and some of its hydrolysis products in various species. Br J Pharmacol 1965; 25: 338-51 48. Schumacher H, Blake D, Gillette G. Disposition of thalidomide in rabbits and rats. J Pharmacol Exp Ther 1968; 160: 201-11 49. Schumacher H, Wilson J, Terapane J, et al. Thalidomide: disposition in rhesus monkeys and studies of its hydrolysis in tissues of this and other species. J Pharmacol Exp Ther 1970; 173: 265-9 50. Eriksson T, Bjorkman S, Roth B, et al. Intravenous formulations of the enantiomers of thalidomide: pharmacokinetic and initial pharmacodynamic characterization in man. J Pharm Pharmacol 2000; 52: 807-17

© 2004 Adis Data Information BV. All rights reserved.

63. Lown K, Bailey D, Fontana R, et al. Grapefruit increases felodipine oral availability in humans by decreasing intestinal CYP3A protein expression. J Clin Invest 1997; 99: 2545-53 64. Hess C, Hunziker T, Kupfer A, et al. Thalidomide-induced peripheral neuropathy: a prospective clinical, neurophysiological and pharmacogenetic evaluation. J Neurol 1986; 233: 83-9 65. Eriksson T, Hoglund P, Turesson I, et al. Pharmacokinetics of thalidomide in patients with impaired renal function and while on and off dialysis. J Pharm Pharmacol 2003; 55: 1701-6

Correspondence and offprints: Dr Steve K. Teo, Celgene Corporation, 7 Powder Horn Drive, Warren, NJ 07059, USA. E-mail: [email protected]

Clin Pharmacokinet 2004; 43 (5)