`DOI IO.I007/s002280 100320
`
`REVIEW ARTICLE
`
`Tommy Eriksson · ven Bjorkman · Peter Hoglund
`Clinical phannacology of thalidomide
`
`ovember 2000 / Accepted in revi ed form: 26 April 200 1 / Published onl ine: 13 Jul y 2001
`Received: 28
`© Springer-Verlag 2001
`
`Abstract Background: Thalidomide has a chiral centre,
`and the racemate of (R)- and (S)-thalidom ide was in(cid:173)
`troduced as a sedative drug in the late 1950s. In 1961 , it
`wa withdrawn due to teratogenicity and neuropathy.
`There i now a growing clinical intere t in thalidomide
`due to it unique anti-inflammatory and immunomod(cid:173)
`ulatory effects.
`tudies
`Objective: To critically review pharmacokinetic
`and briefly review pharmacodynamic effects and tudies
`of thalidomide in consideration of its chemical and
`te(cid:173)
`reochemica l propertie and metabolism.
`Methods: Literature sea rch and computer simulations of
`pharmacokinetics.
`Results: Rational u e of thalidomide is problematic due
`to lack of ba ic knowledge of it mechani m or action
`effects of the epara te enan tiomer and metabolites and
`dose- and concentration-effect relation hip . Due to its
`inhibition or tumour necrosis factor-ex and angiogenesis,
`racemic thalidomide has be n tested with good effect in a
`variety or skin and mucous membrane disorders Cro(cid:173)
`hn 's disease, graft-versus-host disease, complications to
`human immunodeficiency virus and , recently, in multi(cid:173)
`ple myeloma. dver
`reaction are oft n related to the
`edative effect . Irreversible toxic peripheral neuropathy
`and foetal malformation are serious complica tion that
`can be prevented. The re ults of everal published
`pha rmacokinetic studie can be questioned due to poor
`methodology and the u e of non-stereospecific as ays.
`The enantiomer of thalidomide undergo pontaneous
`
`T . Eriksson (18J)
`Hospital Pharmacy, University Hospital , 221 85 Lund, Sweden
`E-mail: tommy.eriksson@apotekct.se
`Tel. : + 46-46- 171265; Fax: + 46-46-159301
`S. Bjorkman
`Hospita l Pharmacy, Malmo niversity Hospital,
`Malmo Sweden
`
`P. Hoglund
`linical Phamrncology,
`Department of
`Divi ion
`f Labornt ry Medicine,
`University Hospita l, Lund, Sweden
`
`hydrolysis and fast chiral interconversion at physiolog(cid:173)
`ical pH. The oral bioavailability of thalidomide has not
`been unequivocally determined, but avai lable data sug(cid:173)
`gest that it is high. Absorption is
`low, with a time to
`max imum pla ma concentration of at lea t 2 h, and may
`al o be do e-dependent; however, that of the eparate
`enantiomers may be fa ter due to higher aqueous olu(cid:173)
`timation of the volume of di tribution i com(cid:173)
`bility.
`plicated by probable hydrolysis and chiral inversion al o
`in peripheral compartments. A value of around I I/ kg is
`however plau ible. Plasma protein binding is low with
`little difference between the enantiomer . Elimination of
`thalidomide is mainly by pH-dependent spon taneous
`hydrolysis in a ll body fluids with an apparent mean
`clearance of IO 1/ h for the (R)- and 21 1/h for the (S)(cid:173)
`enantiomer in adult subjects. Blood concentration of
`the (R)-enantiomer are consequently higher than tho e
`or the (S)-enantiomer at pseudoequilibrium. The mean
`elimination half-l ife of both enan tiomers is 5 h. One
`hydroxylated metabolite has been found in low con(cid:173)
`centrations in the blood. Since both enzyma tic metab(cid:173)
`olism and renal excretion play minor roles in
`the
`elimination of tha lid omide, the ri k of drug interaction
`eem
`to be low.
`onclusions: The intere t in and u e of thalidomide i
`increa ing due to it potential a an immunomodulating
`and antiangiogenic agent. The inter-individual variabil(cid:173)
`ity in distribution and elimination is low. Apart from
`thi , its u e is complicated by the lack of knowledge of
`dose- or concentration-effect relationships, possible
`do e-dependent oral absorption and of course by its
`wel l- known serious adverse effects.
`
`Keywords Thalidomide · Stereoisomers ·
`Pharmacokinetics
`
`Historical perspective
`
`The tory of thalidomide i a tragic chapter in the gen(cid:173)
`erally progressive development of drug therapy. It has
`
`DR. REDDY’S LABS., INC. EX. 1068 PAGE 1
`
`
`
`366
`
`been de cribed a a turning point, an end of innocence
`and the beginning of a new and greater en e of re-
`pon ibility in clinical pharmacology, in clinical inve -
`tigation and in medicolega l relation hip (Curran 1971).
`It also had a profound effect on the drug regu latory
`process (Kelsey 19
`).
`Tha lidomide was synthesised in 1954 by Kunz, a
`chemist at Chem ie Griinenthal GmbH (Kunz 1956).
`the po tu lated anti hi taminic properties for
`Whil t
`which it was synthe ised were weak , it produced marked
`sedation. Doses of tha lidomide in excess of IO g/kg
`fa iled to show lethality in rodents. As a result it was
`regarded as a particularly safe drug. Thalidomide was
`la unch d a a sedative in 1957 and in 1960 it was mar(cid:173)
`in Germany
`keted in 20 countries. In 1960 the sale
`amounted to 14.6 tons and it wa
`o ld free of prescrip(cid:173)
`tion. Report about neuro-pathogenicity after prolonged
`u e of thalidomide began to appear in 1960. In 1961
`McBride and Lenz, two physicians working indepen(cid:173)
`dently of each other, sta rted to susp ct a link between
`birth of children with evere malformations and con-
`umption of thalidomide during the fir t trime ter of
`pregnancy. The drug wa withdrawn from the market
`but it u e had re ulted in more than I 0,000 victim of
`ma lformation with abnormalities uch as phocomelia
`( hort limb ), amelia (ab ence of limb ), ea r eye, heart
`and ga trointe tinal abnormalities (McBride 1961 · Lenz
`1988; Zwingenberger and Wnendt 1996).
`In 1965 there wa a report of remarkable effec t in the
`treatment of the debilitating and di figuring le ion a -
`sociated with erythema nodosum lepro um (ENL), a
`complication of leprosy (Hansen ' disease, Sheskin
`find ing there wa a renewed clinical
`1965). After thi
`intere t in thalidomide, due to its ant i-inflammatory and
`immunomodula ting effects (for reviews see Schuler et al.
`1995 and Zwingenberger and Wnendt 1996). Recently
`the ood a nd Drug dministration of the Un ited States
`approved thalidomide for use in patients with ENL
`(Nightinggale 1998).
`
`Chemistry and chirality
`
`Thalidomide (ix-phtha limidoglu tarimide, Fig. I) ha a
`ch iral centre and i used a a racemate ( I: I mixture) of
`dextrorotatory ( R)- and levorotatory (S)-thalidomide.
`The aqueous solubility of racemic (rac) thalidomide has
`been reported to be about 50 µg / ml , with a five-t imes
`
`0 CX"-k-<--o
`
`0 0 NI
`
`\
`(R)-Thotrdomidc H
`
`H
`I
`
`0 o
`
`CGNf-~{~0
`
`0
`(S)-Tholidomlde
`
`Fig. I Stereochemica l structures of (R)- and (S)-thalidomide
`
`higher olubility of the eparate enantiomer (Hague and
`Smith 1988; Krenn et al. 1992; Eriksson et al. 2000a).
`Thalidomide i
`rapidly degraded by spontaneou
`hydrolysi
`in any aqueous medium at phy io logical pH
`(Schumacher et al. 1965a; Eri ksson et al. 1992, 1997;
`Boughton et al. 1995; Huupponen et al. 1995; Lyon
`et al. 1995). All the substituted amide bonds are sensi(cid:173)
`tive
`to hydroly i and a t pH 7.4 twelve hydrolysi
`products are formed by splitting these bonds. From
`pH 6 to pH 7 only the phtha limide ring undergoe
`cleavage, whereas at pH 7 and above the glutarimide
`moiety also undergoes hydrolysis (Schu macher et al.
`1965a). The rates of inversion and hydrolysis of th e
`enantiomers increase with pH in the interval 7.0- 7.5
`(Eriksson et a l. 1998a).
`Proper hand li ng of solution and biological samples
`of thalidomide and its enantiomers is crucial to avoid
`hydrolysi and chiral in version. Various technique have
`been described for handling of plasma samples, such as
`acidification with HCI (Boughton et al. 1995) or fast
`chi ll ing and tran portation for centrifugation and
`tor(cid:173)
`ing of plasma at - 25°C {Lyon et al. 1995). These meth(cid:173)
`od do not completely inhibit hydroly i of thalidomide
`( ome
`10% degradation i reported) and do not ad(cid:173)
`the problem of racem i ation if one wishes to
`dre
`mea ure the concentrations of the eparate enantiomers
`(Erik on et al. 1997). In the latter ca e both procedure
`will probably be unacceptable, since enantiomeric in(cid:173)
`version i twice a fast as hydrolysi in blood a well a in
`plasma (Eriks on et al. l 998a). In contrast, with the u e
`of the protocol outlined in the appendix there was no
`detectable degradation or racemisation in blood/buffer
`mixture stored at - 25 °C for 75 days and 100 day ,
`respectively (Eriksson et al. 1997).
`
`Pharmacokinetics
`
`There is fast chiral interconversion between the enanti(cid:173)
`omer of thalidomide after intravenous and oral ad(cid:173)
`mini tration of the epara te enantiomers as hown in
`two ubject from our studie (Fig. 2). Since thalidomide
`is in reality two di fferent molecules with differen t phar(cid:173)
`macokinetic and pharmacodynamic profiles, pharmac(cid:173)
`okinetic data
`hou ld be given
`for
`the separate
`enantiomers. However, since rac-thalidornide, in cap(cid:173)
`sules or tablets, is the only form of thalidomide in clin(cid:173)
`ical use, and
`ince our group alone ha performed
`human studies on the enantiomers, a ummary of studies
`in which calculated pharmacokinetic parameters are
`based on total thalidomide concentrations will also be
`presented.
`
`Oral absorption of the racemate
`
`for ab orption of oral
`Pharmacokinetic parameter
`in
`dose of me-thalidom ide have been determined
`healthy volunteer (Beckmann and Kampf 1961 ; Green
`
`DR. REDDY’S LABS., INC. EX. 1068 PAGE 2
`
`
`
`Fig. 2 Blood-concentrntion
`curves of ob erved (R)-thali-
`domide (filled symbols) and
`(S)-thalidomide (open :,y m bols)
`and filled concentra ti ns
`(c11nes) after oral ( I mg/kg)
`or intravenous (i.v., 50 mg over
`60 min) administration of the
`indicated enantio mer to two
`subjects
`
`e
`' "'
`:I.
`c
`-~
`~ 0.1
`c •
`<>
`C
`0 u
`
`0.01
`
`e
`' "'
`::x.
`c -~ E Od
`c • <>
`
`C
`0
`0
`
`0.01
`
`•
`
`4, oral odminislrolion
`
`(R}-tholidomide
`
`Subject 19, i.v. Infusion
`
`(R)-lholldomide
`
`367
`
`0
`
`Subject 4, oral odministro!ion
`
`0
`
`(S)-thalidomide
`
`Subject 19, i.v. infusion
`
`(S)-lholidomlde
`
`0
`
`6
`
`12
`Time, hours
`
`18
`
`24
`
`' . I.I.I ' ' I I'
`12
`6
`0
`Time , hours
`
`18
`
`24
`
`hen et a l. 19 9; Boughton et al.
`and Benson 196 1;
`1995; Trapnell et al. 1999; Teo et al. 1999, 2000b; Cel(cid:173)
`gene Corp. 2000), in patien ts suffering from graft-ver(cid:173)
`sus-ho t
`disea e
`(GVHD)
`after
`bone marrow
`transplantation (Heney et al. 1991 ), in patients with
`human immunodeficiency virus (H IV) infection (Pisci(cid:173)
`telli et al. 1997;
`oormohamed et al. 1999), in elderly
`patien ts with prostate cancer ( igg et al. 1999), in pa(cid:173)
`tients with leprosy (Celgene Corp. 2000) and in patients
`with gliomas (Fine et al. 2000). In the e studies, the
`pharmacokinetics describing total tha lidomide concen(cid:173)
`trations wer pres nted. Their results are summaris d in
`Table I. However, as shown in Table 2, methodological
`weakne e ca t doubt on ome of the finding . In ad(cid:173)
`dition to thi , we have pre ented phannacokinetic data
`for the eparate enantiomer of thalidomide after oral
`administration of the racema te to healthy volunteers
`(Eriksson et a l. 1995).
`Table I demon trate
`that the different studies on
`oral absorption of a low dose (typica lly I 00 mg) of rac(cid:173)
`thalidomide in healthy volun teers have yielded con i -
`tent results. Absorption i slow, with time to maximum
`plasma concentration (tmax) normally at 2-4 h and
`maximum plasma concentration (C.,iax) around I ~tg/ml.
`With larger doses, the poor solubility of thalidomide in
`intestinal fluids may decrea e the rat of absorption
`(
`!gene orp. 2000). Rate of absorption may conse(cid:173)
`quently be do e-dependent. Co-administration of tha(cid:173)
`lidomide with a high-fa t meal wa
`reported to cause
`minor ( < 10% ) change in the ob erved area under the
`curve (AUC) and Cmax value but an increase in lmax to
`approximately 6 h (Teo et al. 2000b).
`
`A large inter-individual variation in plasma con(cid:173)
`centration after oral administration of thalidomide
`was found in pa tients with GVHD (Heney et al. 199 1).
`Two of the four patients differed markedly from the
`profiles obtained in volunteers and in HIV- infected
`patients. The differences were stated to be due mainly
`to a slow absorption rate. The same pattern was de(cid:173)
`th ree of five patients with GVH D and
`scribed
`in
`chronic malabsorption syndrome (Boughton et al.
`1995). Ten hours after an oral thalidom ide doe of
`400 mg the pla ma concentrations in
`the e patients
`wer 0.7, 0.7 and 0.8 ~1g/ml. In two pa tient with nor(cid:173)
`mal ga trointestinal
`function corresponding plasma
`concentration were 5.0 µg/ ml and 6.0 µg/ml. We have
`monitored blood concentrations of tha lidomide in five
`children (0.5- 15 years , 5- 57 kg) with GVHD and po -
`sible chron ic malabsorption yndrome (unpublished
`data). Duplicate blood amples were taken before th e
`morning dose and the measured blood concentration
`were compared with expected concentra tions from a
`computer
`imulation based on our previou data (see
`below). Two of the five chi ldren had lower thalid omide
`concentrations than expected. In one boy (13 years,
`36 kg) the blood concentrations in two samples sepa(cid:173)
`rated by 7 days showed 0. 7 µg/ ml and I. I ~1g/ml when
`3.0 ~tg/ ml was expected. In another boy (9 yea rs, 27 kg)
`the blood concentration was 0.2 µg/ ml when 1.3 µg/ml
`was expected. In contrast to these findings it has been
`suggested that pa tients with leprosy may have an in(cid:173)
`crea ed bioavailability of thalidomide compared with
`healthy ubjects (Celgene Corp. 2000). Thi
`ugge tion
`seems speculative since we have proposed a very high
`
`DR. REDDY’S LABS., INC. EX. 1068 PAGE 3
`
`
`
`368
`
`Table I Pharmacoki neti c parameter for ornl a b orption of rnc-thalidom ide [total , or (R)- a nd (S)-enan ti omer concentrntions], obtai ned
`""'·' maximum plasma concentration, 1,,.0 ,, time to reach Cmax, r 111 half-life, 11,,g
`in different studies (mean ± SD or median, range).
`a bsorption lag time, '"Sa/1 " syrup, "Tropfen ' " drops
`
`o. of patients
`(P) or hea lthy
`volunteers (Hv)
`
`Administration
`fom1 a nd doe
`(mg)
`
`Beckma nn a nd
`Kampf 1961
`
`Gree n and Benson 196 1
`Chen et al. 1989
`
`Heney et al. 199 1
`Bought n et a l. 1995
`Eri k son et al. 1995
`
`5 H v
`7 Hv
`7 H v
`2 H v
`8 H v, fasting
`
`4 P, fas ting
`3 Hv
`6 H v, fasting
`
`Tablet 100
`" Saft" 100
`" T ropfen" 100
`Tablet 150
`Tablet ,
`chewed, 200
`100
`200
`apsule 100"
`
`Pi citelli et a l. 1997
`
`Erik son et al. 1998b
`
`5 p
`4 P
`6 Hv, fasting
`
`Capsule 100
`Capsule 300
`Tablet 100
`
`oormohamed
`et al. 1999
`Figg et al. 1999
`
`Trapnell et al. 1999
`
`14 P, fasting
`
`13 P
`11 P
`9 H v, fasting
`
`Teo et al. 2000a
`
`17 Hv, fa ting
`
`Teo et al. 2000b
`
`Celgene Corp. 2000
`
`13 Hv, fasting
`Fastin g
`High-fa t mea l
`14 Hv
`
`fine et al. 2000
`
`6 P
`34 P
`
`"Recalcula ted from a do e of 1.5 mg g
`
`apsule 100
`Capsule 200
`200
`800
`200, Day I
`200, Day 2 1
`3 Different
`capsules
`
`Tablet 200
`Capsule 200
`Capsule 200
`Capsule 50
`Capsule 200
`Capsule 400
`Capsule 400
`800
`
`Cma:<
`(µg/ml)
`
`0.9
`1.2
`1.8
`1.3
`1.2±0.2
`
`o. 1.5
`1.9 (1.7- 2. 1)
`"(R):0. 7
`(0.6
`.9)
`"(S):0.5
`(0.4
`.6)
`1.2 ± 0.2
`3.5 ± 1.1
`(R):0.5
`(0.4- 0.6)
`(S):0.3
`(0.3
`.4)
`1.15 ± 0.2
`1.9 ± 0.5
`2.0 (1.2- 3.8)
`4.4 (2. 8.4)
`3.2± 1.0
`4. 2±2.0
`2.0 ± 0.5
`2.1 ± 0.5
`1.0 ± 0.3
`1.05 ± 0.3
`2.0 ± 0.4
`2.2 ± 0.5
`0.6 ± 0.3
`1.7±0.5
`2.8 ± 0.8
`3.4 ± 1.8
`4.1 (0. 12)
`
`tma (h)
`
`l1 12abs (h)
`
`t1,g (h)
`
`1.7 ± I. I
`
`0.4 ± 0.6 in 6/8
`
`0-0 7 in 3/6
`
`1.5 ± 0.9
`1.2 ± 0.7
`
`0.3 ± 0.1
`0.3 ± 0.2
`
`0.95 ± 0.96
`1.2 ± 1.2
`
`0.2 ± 0.2 in 7/1 4
`0. 1 ± 0.2 in 4/ 14
`
`4
`2
`2
`4
`4.4 ± 1.3
`
`2
`)
`3 (2
`"( R):4 (3- 5)
`
`"(S):4 (3- 5)
`
`3.4 ± 1.8
`3.4 ± 1.5
`(R):3.5 (3
`
`(S):4 (2- 6)
`
`2.5 ± 1.5
`3.3 ± 1.4
`3.3 (2.0- 7.1)
`4.4 ( 1. 7.1 )
`5.8
`5.
`3.2± 1.4
`3 5 ± 1.6
`3.4 ± 1.4
`6.2 ± 1.9
`4.0 ± I. I
`6.1 ± 2.3
`2.9 ± 1.9
`3.5 ± 2.0
`4. 3± 1.6
`5.7 ± 1.6
`4. 7 ( 1.7- 8.8)
`
`bioavailabi li ty in health y ubject a de cribed below
`(Eriksson et al. 2000a).
`All of the e studies a re small and some of the findings
`might be explained by the methodological problems
`shown in Table 2. However, they do suggest that in
`ome patients th alidomide may have a lower bioavail(cid:173)
`a bility than expected due to general malab orption, non(cid:173)
`linea r a bso rpti on of high do es o r other reaso ns. If a
`high systemic ex posure is crucial for di sease control,
`then therapeutic drug monitoring could be of value.
`
`Oral absorption of the sepa rate enantiomers
`
`Our group investigated the oral abso rption of the sep(cid:173)
`ix healthy vol(cid:173)
`ara te enantiomer and the racemate in
`untee r in a crossover study (Eriksson et al. 1995). The
`dose were 1.0 mg/ kg (R)- or (S)-thalidomide or 1.5 mg/
`
`kg rac-tha lidomide, given a cap ule . There wa a fa ter
`absorption of (R)- and (S)-t halidomide when given
`separately than when they were given as the racemate.
`This was proba bly due to a fa ter di ssol ution of the
`more water-soluble separa te enantiomers. The AUC
`values from thi oral
`tudy were compared with UC
`values from a similar
`tudy in which the tha lidomide
`ena ntiomers were given sepa rately as a n infu ion (Eri(cid:173)
`ksson et al. 2000a). This comparison suggested that the
`oral bioava ila bi lity of the enantiomers is high, possibly
`around 100% a nd 80% for (R)- and (S)-thalidom ide,
`respectively.
`
`Recta l a bsorption of roe-thal idomide
`
`We inve tigated the rectal a bsorption of racemic tha(cid:173)
`lidomide in healthy male volunteers (Eriksson et al.
`
`DR. REDDY’S LABS., INC. EX. 1068 PAGE 4
`
`
`
`Table 2 Background information for judgement of rele ance
`
`f resu lts from stud ies presented in Table I
`
`Adequate precautions
`taken against hydrolysis
`of thalidomide in blood
`or plasma amples
`
`Full description
`of used drugs
`and/or food
`
`Relevant data
`on subjects
`and intake
`
`A ay fully
`described
`and acceptable
`
`369
`
`Beckmann and Kampf 1961
`Gree n and Benson 1961
`hen et al. l 989
`Heney et al. 1991
`Boughton et a l. 1995
`Eriksson et a l. 1995, 1998b
`Pi citelli et al. 1997
`oormohamed et al. 1999
`Figg et al. 1999
`Trapnell el al. 1999
`Teo et al. 2000a, 2000b
`Celgene Corp. 2000
`Fine et al. 2000
`
`?
`?
`
`0
`
`?
`Yes
`Ye
`0
`Yes
`0
`Ye
`Yes
`?
`Yes?
`
`0
`0
`Yes
`Yes
`0
`Yes
`0
`Yes
`0
`Yes
`Yes
`0
`0
`
`0
`0
`Yes
`Yes
`0
`Yes
`0
`Yes
`0
`Yes
`Yes
`0
`0
`
`Yes
`Ye
`Yes
`0
`Ye
`Yes
`Yes
`Yes
`Yes
`Yes
`Yes
`0
`Yes
`
`2000b). The drug was micronised, and hard fat sup(cid:173)
`positories a nd eldexomer rectal gel were te ted com(cid:173)
`pared wi th tablet produced by Griinenthal GmbH
`(Stolberg Germany). The rectal ab orpt ion was slow
`and varia ble. The mean bioavailability relative to oral
`administra tion was below 40%. We concluded that
`rectal admini tration i not uitab le for clinical u e.
`
`Di tribution and elimina tion data referring
`to tota l concen tration of thalidomide
`
`Pharmacok inetic parameters for distribution and elimi(cid:173)
`nation of tha lidomide are availab le from everal sources
`( hen et a l. 19 9; Piscitelli et al. 1997; Figg et al. 1999;
`oormohamed et al. 1999; Trapnell et al. 1999; elgene
`Corp. 2000; Fine et al. 2000; Teo et al. 1999, 2000b) and
`are presented in Table 3. In most stud ies, thalidomide
`showed a n apparent vo lume of distribution after oral
`administra tion (V / ) of approximately I I/ kg body(cid:173)
`weight. Va lue of appa rent oral clearance (CL/ F) a re
`tudie . However, the e
`al o fai rly consistent betwe n
`number repre ent a compo ite of the di stribution and
`elimination of both ena ntiomers (where distribution
`proce e also include rever ible inversion to the other
`enan ti omer, see Eriksson et al. 2000a) and also assume
`complete bioavailabili ty. They are therefore very "ap(cid:173)
`parent" in nature. Thalidomide i eliminated almo t
`exclu ively by spontaneou hydrolysis in vivo. Hepatic
`metabolism a nd renal excretion play very minor roles in
`anima ls (Beck man n 1962; Williams et al. 1965; Schum(cid:173)
`acher et al. 1965b, 1968, 1970) and in man (Williams
`et al. 1965; Chen et al. 1989; Teo et al. 2000a). Animal
`experiments demon trated that hydrolysis takes place in
`plasma and in all examined tissues (Schumacher et al.
`1965b, 1968, 1970). From in vitro findings we have also
`uggested tha t hydrolysi would occur in the entire dis-
`tribution pace of the drug (Erik on et al. 1998a).
`As shown in Table 2 the handling of the blood
`amples in the fir t
`tudy (Chen et al. 1989) can be
`
`que tioned. An additional problem in the as es ment of
`di tribution and elimination data is the blood- am pling
`protocol, with frequent sam ple
`taken over 12 h fol(cid:173)
`lowed by only a
`ingle sample after 24 h. The range of
`apparent
`terminal half-lives of thalidomide
`in
`the
`subjects was 3.0- 14.6 h. This might reflect experimental
`uncertainty
`rather
`than
`actual
`in ter-individual
`variation. The findi ng of a low urinary excretion eern
`more reliable. Urine is normally slightly acidic a nd
`thalidomide wo uld therefore be protected from hydro(cid:173)
`ly is in the samples. Moreover, in a previou study the
`amount excreted unchanged in the urine was I, 2.2
`and 1.8% in man , rats and dogs , respectively (Will iam
`et al. 1965).
`In the study o f Pi citelli et al. ( 1997) the sampling
`schedule was better, but only three samples were taken
`after the 6-h sample (at 23, 27 and 31 h) and the range of
`the terminal half-Jives was also very wide, 3.7- 11 .5 h.
`The same sampling schedule was used by Figg et al.
`( 1999), yielding a simi lar wide range of measured ter(cid:173)
`mina l half-li ves, 2.0- 18.3 h after a 200 mg dose a nd 4.9-
`55.4 h after 800 mg. Similar problems a ppl y for the
`tud y by Fine et al. (2000) with half-live va ryin g from
`2.7 h to 27.9 h after a single oral dose of 800 mg a nd
`being 0. 75- 31.9 h after multiple dai ly do es of 800-
`1200 mg.
`The variation in half-lives wa generally not as pro(cid:173)
`nounced in the later tudie ( oo rmohamed et a l. 1999;
`Trapnell et al. 1999; Teo et al. 1999, 2000b), in which the
`blood sampling can be judged to have been appropriate.
`The Thalomid product monograph (Celgene Corp. 2000)
`reports similar tem1 in al half-lives at three dose levels (50,
`200 and 400 mg) of thalidomide. Slow dissolution of
`thalidomide in the intestine can, however, apparently
`give rise to flip-flop pharmacokinetics, i.e. the observed
`termina l half-l ife represe nts absorpti on instead of elimi(cid:173)
`nation of tha lidomide. Th is was observed by Figg et al.
`(1999) for a high dose (800 mg) of thalidomid e and al o
`by Teo et al. ( 1999, 2000b) for one of the three studied
`products (a a con equence, the calculations of V / are
`
`DR. REDDY’S LABS., INC. EX. 1068 PAGE 5
`
`
`
`370
`
`Table 3 Pharmacokinetic p,trnmeters for distribution a nd elimina tion of total concentration or the separate enantiomer of thalid mide,
`obtained in different studies. VJF volume of distribution, CL/F apparen t oral clearance, 1111 half-life, MRT mea n residence time,
`i. 11. int ravenous
`
`Route, dose
`
`V / F (I)
`
`CL/ F (1/h)
`
`Renal
`CL/ F (1/ h)
`
`1112 (h)
`
`M RT (h)
`
`hen et al. 1989
`Eriksson et al. 1995
`
`Piscitelli et al. 1997
`
`oormohamed et al. 1999
`
`Figg et al. 1999
`
`T rapnell et al. 1999
`
`Teo et al. 1999
`
`Celgene Corp. 2000
`
`Oral , 200 mg
`Oral , I mg/ kg
`
`Oral 100 mg
`300 mg
`Oral 100 mg
`200 mg
`Oral 200 mg
`800 mg
`Oral 200 mg
`
`Oral, 3 Different
`capsules, 200 mg
`
`Oral 50 mg
`200 mg
`400 mg"
`
`121 ± 45.4
`(R):48 ± 8.8
`(S):72 ± 14
`8 ± 13
`78 ± 22
`70 ± 16
`83 ± 35
`67 ± 34
`166 ± 84b
`
`9 ± 26
`77 ± 13
`(240 ± 79)b
`
`10 ± 2.0
`
`0.08 ± 0.03
`
`9.2 ± 1.2
`7.8 ± 1.8
`10.4 ± 2. 1
`10.8 ± 1.7
`7.4 ± 2.0
`7.2 ± 2.9
`5.4 ± 1.9c
`4.1 ± 2.0c
`10.5 ± 2. 1
`10 ± 1.4
`11.4 ± 3.05
`
`Fine et al. 2000
`
`Oral 800- 1200 mgd
`
`(146 ± 92}°
`(124 ± 73)°
`
`13.3 ± 7.8
`12.6 ± 6.6
`
`Teo et al. 2000b
`
`Oral 200 mg
`
`Eriksson et al. 2000a
`
`1-h i.v. infusion
`50 mg
`
`(R):10 ± 2. lr
`(S):21 ± 4.6r
`
`8.7 ± 4.1
`4.7 ± 0.4
`
`6.5 ± 3.4
`5.7 ± 0.6
`4.6 ± 1.2
`5.3 ± 2.2
`6.5 ± 3.8
`18 ± 14
`6.7 ± 1.7°
`6.8 ± 3.1 °
`6.2 ± 2.6
`5.4 ± 1.3
`15±6.0
`5.5 ± 2.0
`5.5 ± 1.4
`7.3 ± 2.6,
`6.8 ± 1.2
`8.3 ± 6.0
`8.3 ± 7.1
`13.5 ± 6.ss
`5.8 ± 1.7s
`5.1 ± 1.0 s
`4.7 ± 0.5
`
`21.4 ± 8.2 s
`10.4 ± 2.3s
`10.9 ± 2.0 s
`R:4.7 ± 0.7,
`:3.9 ± 0.6
`
`• Hea lt hy subjects and patients, respectively
`b ot a va lid estimate, due to flip-flop phamiacokinetics (see tex t)
`0 During 21 days of treatmen t, day I and day 21 , re pectively
`dAfter a single dose of 800 mg, and after multiple daily doses of
`800-1200 mg, respectively
`
`0Possibi lity of flip-flop pharmacokinetics after these high doses
`rApparent clearance from the central compartment, see Eriksson
`et al. 2000a for deta ils
`STablet fas ting, capsule fa sting a nd capsule after a high-fat meal,
`respectively
`
`not va lid in these cases). The stud y by Fi ne et a l. (2000)
`also seems to suffer from these problems.
`We found a lower inter-individual va riability in half(cid:173)
`li ves than any of the other studies (Eriksso n et al. 1995,
`2000a). From a mecha nistic point of view, a v ry low
`variability is expected.
`Since, as discuss d a bove, only pH a nd temperature
`affect the rate of hydroly i of tha lidomide it should be
`very simila r between individuals. Provided that hydro(cid:173)
`ly i takes place uniformly in the body one wo uld even
`expect that the in vivo half-life should be clo e to the
`va lue obtained with incubation
`in blood or pla ma in
`vitro at 37 °C, pH 7.4, as well a
`to half-lives in an imal
`tudie . Thi has al o been confirmed experimentally; the
`half-li fe of thalidomide was virtua ll y the sa me in buffer
`a t pH 7.4, in human blood a nd plasma and in rabbit
`li ver homogenates as in vivo (Eri ksson et al. 1995,
`1998a, 1998b). It also corresponded well with apparent
`t rminal half-life in rats rabbits a nd monkeys (Schum(cid:173)
`acher et al. 1968, 1970).
`In a multiple-do e
`tudy, 200 mg thalidomide was
`give n daily for 21 days to ten hea lthy female patients.
`The pharmacokinetic profile were similar on the fir t
`a nd the last days of dosing, which indicates that tha(cid:173)
`lidomide doe not induce or inhibit its own metabolism
`
`(Tra pnell et a l. 1999). This study also demo nstrated a
`lack of drug-drug interaction between thalid omide a nd
`ethinyl estrad iol or norethindrone.
`
`D istributi o n a nd elimi nation of the ena ntiomer
`
`We have found that the enantiomer of tha lidomide a re
`not extensively bound to blood or pla ma components
`(95%
`(Eriks o n et al. 1998a). The geometric mean
`co nfidence limit ) of pla ma protei n binding were 55 %
`(53- 58%) a nd 66 % (63- 69 %), re pectively, fo r (R)(cid:173)
`and (.$)-thalid omide. The corre ponding blood :pla ma
`di tribution ratios were 0.86 (0.84
`.89) and 0.95 (0.92-
`0.98). Serum albumin, and to a Jes er extent human
`plasma, catalysed the inversion but not the degradatio n
`at pH 7.4. Thus, we suggested that chira l inversion
`takes place chiefly in the circulation and in extravas(cid:173)
`cular (in tersti tial) sites with a high concentration of
`alb umin while it is slower in more peripheral sites of
`di tribution. Rates of hydrol ysis, however, were
`apparently o nly dependent on pH , giving us a rationa le
`for the ugge tion that hydrol y is would take place
`more uni formly in the entire di tribu tion pace of th e
`dru g.
`
`DR. REDDY’S LABS., INC. EX. 1068 PAGE 6
`
`
`
`Table 4 Releva nt clinical
`pharmacokinetic data (median
`values) for (R)-and (SJ-thalido(cid:173)
`mide. Volume refer to central
`vol umes after intravenous infu(cid:173)
`sions (Eriksson el al. 2000a).
`C,,,,,x maximum plasma con(cid:173)
`centration, lmax tin1e to Cma.x,
`GHVD graft-vers us-host dis(cid:173)
`ease, CL,,p1, apparent clearance,
`t 111 ha lf-life, M RT mean resi(cid:173)
`dence time, F ora l bioavailabil(cid:173)
`ity, Ve ce ntral volume
`
`bsorption"
`
`Distribution
`
`Elimination
`
`Interactions
`
`371
`
`Cm .. ,: (R) 0.6 ,tg/ml, (S) 0.4 ,1 ml
`tmax 4 h
`F: Pro babl y high (80- 100% at low doses)
`Presumably variable and delayed at high doses and in some patient with GVHD0
`Ye: (R) 18 I, (S) 24 I
`Plasma protein binding: ( R) 56%, (S) 63%
`Bl ood: plasma distribution ratio ( R) 0.84, (S) 0.96
`Metaboli m: spontaneou degradation to 12 hydrolysis product b
`Low concen trations of hydroxylated metabolites (low ng range)
`o induction or inhibition of its own metabolismd
`Urinary excretion: pre umably around I o;.c
`Lapp: (R) 10 1/h, (S) 21 1/ h
`MRT: ( R) 4.7 h, (S) 3.9 h
`1112 5 h
`ffects of decreased hepa tic or renal function: unlikely due to mechanism
`of eliminatio n
`Affecting distribution: unli kely due to low binding to plasma and blood
`components
`Affecting elimination : unlikely due to mechanism of elimina tion. Shown
`not to occur wi th ethinyl estradi ol or norethindroned
`
`"After an oral dose of JOO mg m e-thal idomide as tablets or capsu les to fasting healthy volunteers, with
`a weight of approxima tely 70 kg
`All data from Eriksson et al. except
`bSchumache r et al. 1965a
`c hen et al.19 9
`dTrapnell et al. J 999
`<Bought on et a l. 1995 and Heney et al. 199 1
`
`for di tribution and
`Pharmacokinetic parameter
`elimina tion of the ena ntiomer of tha lidomide are
`ava ilable from our two human
`tudie (Erik on et al.
`1995, 2000a), which are presented in Table 3. The
`(R)-enantiomer predominated at pse udoequilibrium
`[(R) /(S) ratio 1.7] irre pective of administered enantio(cid:173)
`mer or route. There wa excellent agreement between the
`fitted terminal half-lives in the oral and the intravenous
`tudy (an example of this i
`hown in Fig. 2). This
`unequivocally confirm
`that
`the
`terminal half-lives
`found after oral admini tration repre ent elimination
`and not absorption of thalidomide. The volume of the
`central compartment and apparent clearance were
`than
`for
`(S)-thalidomide
`ignificantly higher
`for
`(R)-thalidomide.
`
`Formation of hydroly i products and metabolites
`
`Twelve hydrolysis products have been identified in hu(cid:173)
`man and a enzymatic hydroxylation is theoretically
`pos ible on five different carbon atoms more than 100
`meta bolite a nd degradation product
`(including ste(cid:173)
`reoisomers) could be formed (Schumacher et al. 1965b).
`We detected 5'-hydroxythalidomide in low concentra(cid:173)
`tions in plasma from eight healthy ma le volunteers who
`had received 100--200 mg thalidomide orally. However,
`5-hydroxy-, 5,6-dihydroxy- or 4,5-dihydroxythalidomide
`could not be found usi ng high-performance liquid
`chromatography (HPLC) with a detection limit of
`1- 2 ng/ ml. 5-Hydroxy- and 5'-hydroxy- but not 4,5-
`dihydroxy- or 5,6-dihydroxy-thalidomide could be
`identified after in vitro incubation of tha lidomide with
`huma n S9 liver homogenates (Erik on et al. 1998b).
`
`Teo et al. (2000a) found 5-hydroxythalidomide (mo tly
`concentrations below the limit of quantification, 50 ng/
`ml) but not 4- or N-hydroxythalidomide in urine from
`patients with leprosy who had received thalidomide as a
`single 400-mg dose. o metabolites could be detected in
`plasma , however, with a much higher limit of detection
`than in our study. They also
`tudied metabolism of
`thalidomide in vitro , by incubation with pooled micro-
`ome
`co ntaining cloned human cytochrome P450
`isoenzymes. It wa concluded that thalidomide doe not
`undergo
`ignifica nt me ta bolism by human cytochrome
`P450 and that clinically important interactions between
`thalidomide and drugs that are metabolised by this
`enzyme system are therefore unlikely. A summary of the
`relevant clinical pharmacokinetic data for the enantio(cid:173)
`mers of
`thalidomide, after administration of
`the
`racemate, is presented in Table 4.
`
`Pharmacodynamics
`
`Mechani m of action
`
`The molecul a r mechanism for the effects of tha lidomide
`still remains unclear (Giinzler 1992; Calabrese and
`leischer 2000). Multiple anti-inflammatory and immu(cid:173)
`nomodulatory effects have been shown both in vivo a nd
`in vitro (for reviews see T seng et al. 1996 and Koch
`1986). After the finding that thalidomide reduces tumour
`necrosis factor (TNF)-a production by enhancing the
`degradation ofTNF-a mRNA (Moreira et al. 1993) with
`other cytokines remaining unaffected (Sampaio et al.
`1991 ), evera l studies have focused on the effect of tha(cid:173)
`lidomide on this cytokine. Elevated erum T F-a level
`
`DR. REDDY’S LABS., INC. EX. 1068 PAGE 7
`
`
`
`372
`
`including