DRUG DISPOSITION
`
`Clin. Pharmacokinet. 1997 Feb; 32 (2): 120-131
`0312-5963/97/00J2{) 1201$06.0010
`
`© Adis International Limited. All rights reserved.
`
`The Clinical Pharmacokinetics
`of Cladribine
`
`Jan Liliemark
`Department of Oncology, Karolinska Hospital, Stockholm, Sweden
`
`Contents
`Summary
`1. Bioanalysis. . .
`2. Bioavailability .
`2.1 Oral Administration .
`2.2 Subcutaneous Administration .
`2.3 Rectal Administration .... .
`3. Distribution . . . . . . . . . . . . . .
`3.1 Penetration of the Blood-Brain Barrier
`4. Metabolism . . . .
`4.1 Catabolism
`. . . . . . . . . . .
`4.2 Bioactivation . . . . . . . . . .
`4.3
`Intracellular Pharmacokinetics
`4.4 Plasma/Cell Concentration Relationship
`5. Pharmacodynamic Relationships
`6. Excretion ........ .
`7. Interaction ....... .
`8. Interspecies Differences .
`9. Conclusion ....... .
`
`120
`121
`122
`122
`123
`123
`123
`123
`124
`124
`124
`124
`125
`125
`127
`127
`128
`129
`
`Summary
`
`Cladribine is a new purine nucleoside analogue with promising activity in
`low-grade lymphoproliferative disorders, childhood acute myelogenous leukae(cid:173)
`mia and mUltiple sclerosis. Reversed phase high performance liquid chromatog(cid:173)
`raphy and radioimmunoassay have been used for the analysis of the plasma
`pharmacokinetics of cladribine. The major (inactive) metabolite in plasma, chloro(cid:173)
`adenine, can only be detected by liquid chromatography.
`The oral bioavailability of cladribine is 37 to 51 %, and that of subcutaneous
`administration is 100%. The terminal half-life varies from 5.7 to 19.7 hours and
`the apparent volume of distribution from 54 to 357 Llm2
`. The concentration in
`the cerebrospinal fluid is 25% of that in plasma in patients without central nervous
`system disease; in patients with meningeal disease, the cladribine concentration
`in the cerebrospinal fluid exceeds that in plasma.
`Cladribine is a prodrug and needs intracellular phosphorylation to active
`nucelotides. The intracellular concentration of these metabolites is several hun(cid:173)
`dred-fold higher than that of cladribine in plasma and they are retained in leukae(cid:173)
`mia cells with half-lives between 9 and >30 hours depending on diagnosis and
`
`Petitioner TWi Pharms., Inc.
`EX1009, Page 1 of 12
`
`

`

`Clinical Pharmacokinetics of Cladribine
`
`121
`
`sampling schedule. There is no correlation between the plasma concentration of
`cladribine and that of the intracellular metabolites.
`The renal clearance of cladribine is 51 % of total clearance and 21 to 35% of
`an intravenously administered dose is excreted unchanged in the urine. Pretreat(cid:173)
`ment with cladribine increases the intracellular accumulation of the active me(cid:173)
`tabolite of cytarabine, cytosine arabinoside 5'-triphosphate, by 36 to 40%.
`
`The first report on the synthesis and antileukae(cid:173)
`mic effect of cladribine (2-chloro-2/-deoxycytid(cid:173)
`ine, CdA) was published in 1972.[1] Later it was
`discovered that severe immunodeficiency in children
`was, in a fraction of patients, caused by deficient
`adenosine deaminase (ADA).[2] Deoxyadenosine ac(cid:173)
`cumulates in plasma and dATP in cells with high
`deoxycytidine kinase (dCK) activity. Such pertur(cid:173)
`bations of the deoxyribonucleotide pools leads to
`DNA-strand breaks, poly(ADP)ribosyl transferase
`activation, consumption of NAD, ATP depletion
`and loss of viabilityP] Cladribine and other C-2
`halogenated purines (e.g. fludarabine) are resistant
`to ADA due to protonation at N-7 instead of N-6[4]
`which prevents hydroxylation and deamination at
`N-6. Cladribine 5'-triphosphate, with similar toxic
`effects to those of dATP, accumulates in dCK-rich
`tissues; treatment with cladribine can, therefore,
`mimic ADA-deficiency.
`Due to the lack of a solid patent, the develop(cid:173)
`ment of the drug for clinical use was severely
`slowed. Thanks to the efforts of Dennis Carson and
`Ernest Beutler at Scripps Clinic, La Jolla, cladrib(cid:173)
`ine was eventually taken through preclinical and
`early clinical testing.l5-7] Under the Orphan Drug
`Act, cladribine was licensed as Leustatin® in 1994
`and has emerged as one of the more important
`drugs in the therapeutic armament against lympho(cid:173)
`proliferative disorders.[8] The use of cladribine in
`children was investigated independently at St Jude
`Children's Hospital, Memphis, USA.[9] While the
`metabolism and mechanism of action of cladribine
`was elucidated early, the clinical pharmacokinetics
`have not been delineated until recently. The phar(cid:173)
`macokinetic data has, however, grown steadily
`during recent years.
`Cladribine is the drug of choice for the treatment
`of hairy cell leukaemia.l 10, II] It has definite activity
`
`in chronic lymphocytic leukaemia[12-15] and low(cid:173)
`grade non-Hodgkin's lymphoma[15-18] although the
`exact role of cladribine in the treatment of these
`diseases is still a matter of some controversy. The
`use of purine analogues in the treatment of low(cid:173)
`grade lyphoproliferative disorders was recently re(cid:173)
`viewed.l19] Responses are also seen in acute my(cid:173)
`elogenous leukaemia in children[20] and in psoriatic
`arthritis.[21] A randomised, double-blind, cross-over
`trial showed impressive activity in patients with
`chronic progressive multiple sclerosis,l22,23] In exper(cid:173)
`imental systems, cladribine potentiates the immuno(cid:173)
`suppressive effect of cyclosporin and has potential
`in the treatment of transplant rejectionp4]
`
`1. Bioanalysis
`
`The plasma pharmacokinetics of cladribine
`have been studied during continuous infusion and
`after intravenous short infusion, subcutaneous in(cid:173)
`jection, oral, and rectal administration. The concentra(cid:173)
`tion of cladribine has been determined with liquid
`chromatography[25,26] and radioimmunoassay. [20,27]
`Liquid chromatography has the advantage of iden(cid:173)
`tifying the catabolite 2-chloroadenine (CAde)
`while radioimmunoassay can be somewhat more
`
`Fig. 1. The chemical structure of cladribine (2-chloro-2'(cid:173)
`deoxyadenosine).
`
`© Adis International Limited. All rights reserved.
`
`Clin. Pharmacokinet. 1997 Feb; 32 (2)
`
`Petitioner TWi Pharms., Inc.
`EX1009, Page 2 of 12
`
`

`

`122
`
`Liliemark
`
`0.14 mg/kg
`intravenously
`EEll
`
`0.28 mg/kg
`orally
`
`+
`
`0.14 mg/kg
`subcutaneously
`
`+
`
`"0_
`
`1000
`Ql c:
`:c '':::: :J
`$"0 uE
`oS 100
`c: ttl
`.g ~
`~.!Q
`c:o.
`8 .s
`c:
`0
`U
`
`10
`0
`
`0
`
`12
`
`24
`
`36
`
`48
`
`60
`
`72
`
`Time (h)
`Fig. 2. The plasma concentration of cladribine after intravenous [area under the concentration-time curve (AUG) = 791 nmollL·
`hI, oral (AUG = 878 nmollL· h), and subcutaneous (AUG = 709 nmol/L· h) administration of cladribine (from Liliemark et al.,I30J with
`permission) .
`
`sensitive (detection limit I vs 0.2 nmol/L). The
`comparison of pharmacokinetic data for cladribine
`between investigators is obscured by differences in
`the absorption coefficient (£ = 15.0 X 103 Umol)[1.28]
`used for determination of the cladribine concentra(cid:173)
`tion in standards.£5] In the early clinical and phar(cid:173)
`macokinetic studies (before 1993) the absorption
`coefficient for chloroadenine (£ = 12.6 X 103
`Llmol) was used. Therefore, the dose and plasma
`concentration in these studies were overestimated
`by 15%.
`
`2. Bioavailability
`
`2.1 Oral Administration
`
`There are 4 reports on the oral bioavailability of
`cladribine.£29-32] The oral bioavailability in these
`trials varies from 37 to 51 % when the saline solu(cid:173)
`tion for intravenous administration is given to pa(cid:173)
`tients to drink.[29.30,32] Thus, when administered
`orally at about twice the intravenous dose, the areas
`under the concentration-time curve (AUC) are sim(cid:173)
`ilar (fig. 2).
`When compared with 2-hour intravenous infu(cid:173)
`sion (0.12 mg/kg '" 5 mg/m2) the maximum con(cid:173)
`centration was slightly higher but the peak briefer
`after oral (0.24 mg/kg '" 1 0 mg/m2) or subcutane(cid:173)
`ous administration (0.12 mg/kg) [142 nmollL after
`intravenous infusion vs 165 nmol/L after oral and
`and 268 nmol/L after subcutaneous administra-
`
`tion].l30] In some other antimetabolites (i.e. cytara(cid:173)
`bine and 6-mercaptopurine),[33,34] bioactivation is
`impaired when the drug concentrations achieved
`are above the Michaelis-Menten constant (Km)
`of the bioactivating enzymes. The plasma concen(cid:173)
`tration of cladribine achieved by subcutaneous or
`oral administration is, however, at least one log
`lower than the Km of dCK, the intracellular CdA phos(cid:173)
`phorylation enzymeP5] Thus, the drug concentra(cid:173)
`tions achieved with a standard dose are far from
`saturating the bioactivation.
`The interindividual variability of the bioavaila(cid:173)
`bility is considerable [coefficient of variation (CY)
`= 28%] but the variability of the AUC after oral
`administration is not any greater than that after in(cid:173)
`travenous administration (CY = 38 vs 36%).£30]
`Cladribine is not stable at low pH (pH < 2) .
`However, attempts to increase the pH in the stom(cid:173)
`ach with omeprazole before the oral administration
`of cladribine have not improved the bioavailability
`significantly.(29] Neither was the use of enteric
`coated capsules of any benefit.[30] Concomitant food
`intake slowed and lowered the uptake of cladrib(cid:173)
`ine.(29] However, despite a limited bioavailability,
`oral administration of cladribine has been used suc(cid:173)
`cessfully in the treatment of previously untreated
`chronic lymphocytic leukaemia(36,37] and psoriatic
`arthritispl]
`
`© Adis Internotionallimited. All rights reserved .
`
`Clin. Phonnacokinet. 1997 Feb ; 32 (2)
`
`Petitioner TWi Pharms., Inc.
`EX1009, Page 3 of 12
`
`

`

`Clinical Pharmacokinetics of Cladribine
`
`123
`
`2.2 Subcutaneous Administration
`
`Cladribine has no local tissue toxicity and when
`given subcutaneously, the bioavailability is 100%)30]
`Subcutaneously administered cladribine was used
`in 1 large trial on hairy cell leukaemia. The thera(cid:173)
`peutic results seems to be equivalent with those
`after continuous intravenous infusion,flO]
`
`2.3 Rectal Administration
`
`Based on the impressive improvement of the
`bioavailability of 6-mercaptopurine with rectal
`administration as compared with oral administra(cid:173)
`tion,[38] an attempt has been made to improve the
`bioavailability of cladribine with rectal adminis(cid:173)
`tration. However, cladribine is degraded by bacte(cid:173)
`rial enzymes and the bioavailability by the rectal
`route is poor, only 20%.f39]
`
`3. Distribution
`
`During continuous infusion (5 to 10 mg/m2 per
`24 hours which corresponds to a dose of 0.12 to
`0.24 mg/kg per 24 hours)[40] the concentration of
`
`100
`
`~
`0
`E
`.S-
`ID c:
`:0
`:§
`OJ
`U
`OJ
`E
`
`'" OJ c::
`
`10
`
`cladribine in plasma is lO to 50 nmoIlL,[6,20,41]
`while during a 2-hour infusion of 5 mg/m2 it is 100
`to 400 nmoIlU27,41,42] (fig. 3). There appears to be
`a linear dose/concentration relationship for cladrib(cid:173)
`ine in this dose range (0.2 mg/m2 per hour to 2.5
`mg/m2 per hour) although Marks et aI.l421 reported
`a dose-dependent clearance decreasing from 52.7
`to 26.5 Llh/m2 with increasing dosages from 3.5 to
`lO.5 mg/m2.
`The plasma elimination follows a 2- or 3-com(cid:173)
`partment open model depending on sampling pro(cid:173)
`cedure. The terminal half-life (tIS) is relatively
`long, 7 to 19 hours. The pharmacokinetics of clad(cid:173)
`ribine after intravenous administration are summar(cid:173)
`ised in table I.
`
`3.1 Penetration of the Blood-Brain Barrier
`
`The concentration of cladribine in the cerebro(cid:173)
`spinal fluid is approximately 25% of the plasma
`concentration at dose rates between 0.17 mg/m2/h
`and 2.5 mg/m2/h intravenously in patients without
`known meningeal disease,f43,44,46] The kinetics of
`the cerebrospinal fluid concentration roughly fol(cid:173)
`lows those of the plasma.f431 The cerebrospinal fluid
`concentration increases linearly with doseJ43,46]
`No data are available on the penetration of cladrib(cid:173)
`ine into cerebral tissues. However, several responses
`have been noted in patients with astrocytomas, in(cid:173)
`dicating that the drug is distributed to the brain, at
`least when the blood-brain barrier is damaged. Fur(cid:173)
`thermore, cerebral toxicity was noted in patients
`treated with higher doses of cladribine (0.4 to 0.5
`mg/kg daily for 7 to 14 days) than commonly
`used. l27]
`In 2 papers it has been reported that cladribine
`can indeed have an effect on meningeal disease
`when administered intravenously,f47,48] In 1 patient,
`with a meningeal involvement of Waldenstrom's
`macroglobulinaemia, 'the cerebrospinal fluid con(cid:173)
`centration of cladribine exceeded that in plasma,
`which suggests that meningeal involvement in(cid:173)
`creases the penetration of cladribine into the cere(cid:173)
`brospinal fluid,f47]
`
`18
`
`24
`
`a
`
`6
`
`12
`Time (h)
`Fig. 3. The mean plasma concentration of cladribine after 2
`hours intravenous infusion in 12 patients. The data are fitted to
`a 3-compartment open model. The horizontal line shows the
`steady-state concentration of cladribine in the same patients
`after continuous intravenous infusion of the same dose during
`24 hours. The broken lines represent 1 SO in the population
`(from Liliemark and Juliusson,!411 with permission).
`
`© Adis International Limited. All rights reserved.
`
`elin. Pharmacokinet. 1997 Feb: 32 (2)
`
`Petitioner TWi Pharms., Inc.
`EX1009, Page 4 of 12
`
`

`

`124
`
`Liliemark
`
`Table I. Plasma pharmacokinetics of cladribine
`
`Diagnosis
`
`n
`
`Dose/day Duration Mean AUC
`(mg/m2)
`(h)
`(!lmoi/l. h)
`
`t'f2u mean
`(min)
`
`!1/2~
`(h)
`
`t%y
`(h)
`
`Vdss mean
`(SD) [Um2]
`
`Reference
`
`CI mean
`(SD)
`[Uhlm2]
`
`0.59
`0.76
`0.70
`
`2
`2
`2
`24
`24
`2
`2
`
`0.26-1.47
`0.53-1.14
`
`11.9
`
`1.5
`
`5.5
`12
`Cll, HCl, and NHl
`13
`5.0
`Clland NHl
`4
`5.0
`Cll
`5
`8.9
`Paediatric AMl and All
`8.9
`Paediatric AMl
`25
`3.5-10.5
`Solid tumour
`25
`6-12
`11
`CMl
`a Vd of the peripheral compartment.
`b Dose dependent.
`Abbreviations and symbols: All = acute lymphocytic leukaemia; AMl = acute myelogenous leukaemia; AUC = area under the concentration(cid:173)
`time curve; CI = clearance; Cll = chronic lymphocytic leukaemia; CMl = chronic myelogenous leukaemia; h = hour(s); HCl = hairy cell
`leukaemia; NHl = non-Hodgkin's lymphoma; n = number of patients; t%u = distribution half-life; t'f2~ = elimination half-life; \1;.,y = terminal
`half-life; Vd = volume of distribution.
`
`1.1
`0.7
`
`8
`
`3.1
`
`6.3
`9.9
`13.4
`14.2
`19.7
`5.7
`6.0
`
`368 (216)"
`53.6 (23.7)
`
`41
`43
`25.9 (7.8)
`29
`39.1 (6.8)
`305.1
`36.1
`20
`356.6 (225.2) 39.4 (12.4) 44
`26.5-52.7b
`121
`45
`174
`45.1
`42
`
`4. Metabolism
`
`4.1 Catabolism
`
`CAde is the major metabolite in plasma.[25] The
`higher concentration of this catabolite in plasma
`after the oral administration of cladribine (fig. 4) is
`probably due to the degradation of cladribine by
`gastric acid and the subsequent absorption of the
`metabolite or by hepatic or intestinal first-pass ef(cid:173)
`fect. CAde has no cytotoxic or anti tumour effect at
`the concentrations achieved.
`
`however, seems to be phosphorylated readily to the
`monophosphate but, at least in vitro, the concentra(cid:173)
`tions of the di- and triphosphate nucleotides are
`approximately 7 and 3 times lower than those of
`the monophosphate. [9,50]
`Cytoplasmatic 5'-nuc1eotidase[51] dephosphoryl(cid:173)
`ates and deactivates cladribine 5' -monophosphate.
`The level of this enzyme in tumour cells seems to
`be important for sensitivity to cladribine treat(cid:173)
`ment[51,52] and probably determines the retention
`of cladribine nucleotides in tumour cells.
`
`4.2 Bioactivation
`
`4.3 Intracellular Pharmacokinetics
`
`Cladribine is phosphorylated to its 5'-mono(cid:173)
`phosphate by dCK[7] and deoxyguanosine kinase.l49]
`This latter enzyme is found in mitochondria and a
`high activity is present in samples from brain tu(cid:173)
`mours and melanomas[49] and may be of impor(cid:173)
`tance for the therapeutic effect in these tumours.[46]
`However, deoxyguanosine kinase is a mitochon(cid:173)
`drial enzyme and it is unclear whether the phos(cid:173)
`phorylation of cladribine in mitochondria induces
`the same cytotoxic effects as phosphorylation in
`the cytoplasm by dCK. dCK phosphorylates a
`number of nucleoside analogues and is the rate lim(cid:173)
`iting enzyme in the bioactivation of cytarabine
`(ara-C) and fludarabine, another adenosine deami(cid:173)
`nase resistant purine analogue (fig. 5). Cladribine,
`
`The intracellular pharmacokinetics of the total
`cladribine nucleotide pool have been described in
`hairy cell, chronic lymphocytic, and acute myelog(cid:173)
`enous leukaemias after intravenous, oral, and sub(cid:173)
`cutaneous administration. [31] The intracellular con(cid:173)
`centration of the cladribine nucleotides is several
`hundred-fold higher than the plasma concentration
`of the parent drug (fig. 6). Furthermore, the clad(cid:173)
`ribine nucleotides are well retained in leukaemia
`cells in patients with chronic lymphocytic leukae(cid:173)
`mia with a terminal tYz of around 30 hours when the
`cellular concentration is monitored during 3 to 7
`days (fig. 6). These data support the use of inter(cid:173)
`mittent administration, which has been gaining
`widespread use lately.l53-55]
`
`© Adis International Limited. All rights reserved.
`
`Clin. Pharmacokinet. 1997 Feb: 32 (2)
`
`Petitioner TWi Pharms., Inc.
`EX1009, Page 5 of 12
`
`

`

`Clinical Pharmacokinetics of Cladribine
`
`125
`
`However, the retention of cladribine-nucleo(cid:173)
`tides in vivo in leukaemia cells of patients with
`acute myelogenous leukaemia appears to be some(cid:173)
`what lower. During the first 24 hours after admin(cid:173)
`istration the t~ is 9.0 hours in acute myelogenous
`leukaemia patients as compared to 12.9 hours in
`chronic lymphocytic and 15.1 hours in hairy cell
`leukaemiaPI] Twice-daily administration of clad(cid:173)
`ribine in acute myelogenous leukaemia might,
`therefore, be worthwhile.
`Recently a new method was developed allowing
`a specific determination of cladribine mono- and
`triphosphate in leukaemia cells in vivo. Prelimi(cid:173)
`nary data shows that the pharmacokinetics of the
`mono- and triphosphate are similar. However, the
`relation between the 2 intracellular nucleotides
`seems to vary[56] (fig. 7).
`
`4.4 Plasma/Cell Concentration Relationship
`
`The interindividual differences in the intracel(cid:173)
`lular metabolism of cladribine are important since
`
`there seems to be no direct relationship between
`plasma AUC and intracellular cladribine-nucleo(cid:173)
`tide AUe. Thus, the concentration of the active
`metabolites at the target site, i.e. the malignant cell,
`cannot be predicted from plasma concentrations in
`the individual patient. In contrast, the intracellular
`concentration of cladribine-nucleotides appears to
`depend on both the plasma cladribine concentra(cid:173)
`tion and the activity of dCK in the leukaemia cells
`(fig. 8»)57]
`
`5. Pharmacodynamic Relationships
`
`It was recognised early in the development of
`the drug that the action of cladribine is highly
`schedule dependent. [50,58] Therefore continuous in(cid:173)
`fusion for 7 days was chosen as the preferred mode
`of administration in the early clinical trials,l5] How(cid:173)
`ever, several investigators have shown that inter(cid:173)
`mittent administration of cladribine works as well
`in both hairy cell leukaemia[lO] and chronic lym(cid:173)
`phocytic leukaemia.[l3,14]
`
`Orally
`10 mg/m2
`
`Intravenous
`5 mg/m2
`
`o Cladribine
`• 2-Chloroadenine
`
`•
`
`•
`
`~-----.
`
`o
`
`o
`
`140
`
`120
`~
`0
`E 100
`-S
`Q) c
`:0
`~ 80
`C\l
`C3
`'0 60
`c
`0
`~
`C 40
`Q)
`()
`c
`0
`()
`C\l 20
`E
`'" C\l
`i:L
`
`0
`
`-20
`-10
`
`o
`
`10
`
`20
`
`Time (h)
`
`30
`
`40
`
`50
`
`Fig. 4. The plasma concentration of cladribine and its catabolite, 2-chloroadenine in 1 patient after an intravenous 2-hour infusion (5
`mg/m2) and an oral dose (10 mg/m2) of cladribine (Albertioni et aI., unpublished data).
`
`© Adis International Limited. All rights reseNed.
`
`Clin. Pharmacokinet. 1997 Feb; 32 (2)
`
`Petitioner TWi Pharms., Inc.
`EX1009, Page 6 of 12
`
`

`

`126
`
`Liliemark
`
`1 - - - - - - - - - - - - - - - Fludarabine
`
`Cytarabine
`
`Adenosine
`deaminase
`
`1 - - - - - - - - --
`
`Cladribine
`
`Deoxyinosine .... t----+----:::=:----
`
`~~:)
`
`(pentostatin)
`
`, ...
`
`Deoxyadenosine
`S'-triphosphate
`
`...
`
`Fludarabine
`5' -triphosphate
`
`Cladribine
`S'-triphosphate
`
`...
`
`Cytarablne
`S'-triphosphate
`
`Fig. 5. The intracellular metabolism of nucleoside analogues. Deoxycytidine kinase phosphorylates the nucleoside analogues to their
`5'-monophosphates and adenosine deaminase deaminates deoxyadenosine. When pentostatin inhibits adenosine deaminase an
`excess of dATP is formed. Cladribine and fludarabine are resistant to adenosine deaminase. Cladribine is also a substrate for the
`mitochondrial enzyme deoxyguanosine kinase.
`
`In vitro data suggest that there is no simple re(cid:173)
`lationship between the formation of cladribine tri(cid:173)
`phosphate intracellularly and antileukaemia ef(cid:173)
`fect)591 Nor was there any correlation between the
`AVC of the total cladribine nucleotide pool or
`plasma cladribine concentration and response to
`treatment in chronic lymphocytic leukaemia)571
`On the other hand, a weak (p = 0.028) relation(cid:173)
`ship was found between the plasma cladribine
`AVC and the response in hairy cellleukaemia.[lOl
`However, considering the lack of correlation be(cid:173)
`tween plasma and cellular drug concentrations
`mentioned above, this finding needs to be assessed
`critically and needs to be confirmed before its clin(cid:173)
`ical relevance can be determined.
`
`A weak but significant relationship between the
`activity of dCK in leukaemia cells and response
`was also found by 2 independent groupS)52,601
`Thus, the available data indicates that there are
`indeed unknown factors beyond the bioactivation
`of cladribine which determine the final response.
`They do not allow any conclusion on the usefulness
`of the clinical pharmacokinetics for individ(cid:173)
`ualisation of the treatment. However, the intracell(cid:173)
`ular pharmacokinetics of cladribine provides im(cid:173)
`portant information on the effect of route and mode
`of administration for distribution, bioactivation,
`and retention of the drug at the site of action, the
`tumour cell. Thus, intracellular pharmacokinetics
`are an important tool in decision making on the
`
`© Adis International Limited. All rights reserved.
`
`Clin. Pharmacokinet. 1997 Feb; 32 (2)
`
`Petitioner TWi Pharms., Inc.
`EX1009, Page 7 of 12
`
`

`

`Clinical Pharmacokinetics of Cladribine
`
`127
`
`route and interval of administration in treatment
`protocols_
`
`6. Excretion
`
`About 21 to 32% of intravenously administered
`cladribine is excreted unchanged in the urine
`within the first 24 hours[42,45,61] while renal clear(cid:173)
`ance is 51 % (11 to 85) of the total systemic cladrib(cid:173)
`ine clearance.l44] After oral administration, 25 ±
`21 % of the dose (corrected for bioavailability) was
`excreted unchanged in the urine and 3.8 ± 1-9% as
`CAde. A small amount of CAde is found in plasma
`also after intravenous administration and 1.5 ±
`1-6% of the dose is excreted renally as CAde dur(cid:173)
`ing the first 24 hours (Lindemalm et aI., personal
`correspondence). Since most trial protocols re-
`
`quire that patients have normal hepatic and renal
`function, there are little pharmacokinetic or toxic(cid:173)
`ity data on the significance of impaired organ func(cid:173)
`tion. It is therefore not possible to give any recom(cid:173)
`mendations on dosage modifications in such
`situations.
`
`7. Interaction
`
`The only interaction with cladribine which has
`been well described is with cytarabine. Pretreat(cid:173)
`ment of patients with cladribine increases the in(cid:173)
`tracellular accumulation of cytosine arabinoside
`5'-triphosphate (ara-CTP), the active metabolite of
`cytarabine, by 36 to 40%.[62,63] This effect might
`be due to the inhibition of ribonucleotide reductase
`decreasing deoxyribonucleotide pools, in particu-
`
`10000
`
`1000
`
`:J
`'is
`E
`.S-
`co
`0
`~ 100
`C
`Q)
`t.> co
`0
`()
`
`10
`
`, ,
`
`, '
`"
`,
`
`"
`' ,
`
`, ,
`"
`,
`
`~
`
`0.7 mg/kg over 120h
`
`0
`
`o Cladribine, 2-h infusion
`• Cladribine, continuous infusion
`o Cladribine nucleotides, 2-h infusion
`• Cladribine nucleotides, continuous infusion
`
`'()
`
`0.14 mg/kg for 2h
`
`5
`
`2 1 1 1 1 I)
`
`I
`0
`
`I
`48
`
`I
`96
`
`I
`144
`Time (h)
`
`I
`192
`
`I
`240
`
`Fig. 6. The concentration of cladribine in plasma (bottom of panel) and the concentration of cladribine nucleotides in leukaemic cells
`(top of panel) in 1 patient with chronic lymphocytic leukaemia given 0.6 mg/kg (~5 mg/m2) as a 2-hour intravenous infusions and 4
`weeks later as a continuous infusion of the same dose during 5 days. The intracellular concentration of cladribine nucleotides was
`determined by measuring cladribine after dephosphorylation of the nucleotides by alkaline phosphatase in methonal extracts of
`leukaemic cells (from Liliemark and Juliusson,I31J with permission).
`
`© Adis International Limited. All rights reserved.
`
`Clin. Pharmacokinet. 1997 Feb: 32 (2)
`
`Petitioner TWi Pharms., Inc.
`EX1009, Page 8 of 12
`
`

`

`128
`
`1000
`
`10
`
`o
`
`12
`
`24
`
`48
`
`60
`
`72
`
`36
`Time (h)
`
`Fig. 7. The concentration of cladribine in plasma and the clad rib(cid:173)
`ine nucleotides cladribine 5'-monophosphate, and cladribine 5'(cid:173)
`triphosphate in leukaemic cells in 2 patients after an oral dose
`of 10 mg/m2 (Albertioni et al.,156) with permission).
`
`lar deoxycytidine 5' -triphosphate (dCTP) which is
`an inhibitor of cytarabine (and cladribine) phos(cid:173)
`phorylation by dCK. Although there are no data, it
`can be suspected that other nucleoside analogues
`which are bioactivated by dCK, e.g. fludarabine
`and gemcitabine, can be the subject of a similar
`interaction with cladribine.
`
`8. Interspecies Differences
`
`Pharmacokinetically guided dose escalation has
`been applied successfully to minimising the num(cid:173)
`ber of patients needed for phase I dose finding
`
`Liliemark
`
`studies. Traditionally the dose in phase I trials is
`based on the LD 10 in mice, 10% of the LD 10 in mice
`is used as a starting dose in patients. The dose is
`then escalated arbitrarily according to a Fibonacci
`scheme[641 until the maximum tolerable dose is
`reached.
`The pharmacokinetically guided does escala(cid:173)
`tion concept is based on the assumption that there
`is a stronger relationship between the drug concen(cid:173)
`tration (AVC) and toxic effects in mice and humans
`than there is between dose and toxic effects. [65] The
`AVC at LDiO is established in mice. Then the AVC
`in humans at the starting dose (10% of the LD iO) is
`determined. The subsequent dose escalation is then
`based on the difference between the murine LDiO
`and human AVe.
`Theoretically this concept is useful for a number
`of different anticancer drugs. However, for antime(cid:173)
`tabolites, e.g. fludarabine, the concept would have
`been less useful as the AVC in humans at the
`maximum tolerable dose is 10 times lower than
`that of mice at LDiO.l651 Thus, mice appear to be
`less sensitive to the toxic effects of fludarabine.
`This is also the case for cladribine. The Km of the
`main bioactivating enzyme, dCK, is 10 times lower
`for cladribine in human thymocytes as compared
`with thymocytes in mice, while the V max is sim(cid:173)
`ilar. Human dCK therefore phosphorylates clad(cid:173)
`ribine more efficiently than does the murine en(cid:173)
`zyme.l661
`In the original phase I trial of cladribine, the
`first 2 patients were given 1 mg/kg/day as a contin(cid:173)
`uous infusion for 7 days based on toxicity in
`mice.l61 However, the AVC of cladribine at LDIO
`in mice is about 50 times larger than that in patients
`at maximum tolerable dose.l661 The therapeutic
`effect was impressive but the patients developed
`bone marrow failure and died. In subsequent pa(cid:173)
`tients the dosage was decreased to one tenth of the
`starting dose, 0.1 mg/kg/day for 7 days.l61 This
`dose was also considered to be the maximum toler(cid:173)
`able dose which has been confirmed in more recent
`phase I trials.[421 Information on only the plasma
`pharmacokinetics in mice and humans is therefore
`not useful for the determination of a safe dosage of
`
`© Adis International Limited. All rights reserved.
`
`Clin. Pharmacokinet. 1997 Feb: 32 (2)
`
`Petitioner TWi Pharms., Inc.
`EX1009, Page 9 of 12
`
`

`

`Clinical Pharmacokinetics of Cladribine
`
`129
`
`antimetabolites in humans. This shows that know(cid:173)
`ledge about bioactivation is important when phar(cid:173)
`macokinetically guided dose escalation is em(cid:173)
`ployed in phase I trials with antimetabolites.
`There is also a good correlation between toxic
`effects in vivo in humans and the cytotoxic effect
`of antimetabolites in vitro. Thus, information on
`the cytotoxic effect on human cells in vitro is also
`useful when decisions are made about the rate of
`dosage escalation in phase I trialsJ671
`
`....J
`
`'5
`E
`3
`
`600
`
`500
`
`400
`
`300
`
`200
`
`100
`
`•
`•
`•
`•
`# • •
`•••
`• • ~. I· •
`• Ii! •••
`•
`•
`
`.......
`
`r2 = 0.02
`P = 0.25
`
`•
`
`•
`• • •
`
`•
`
`0+---'---'---'---'---'---'---'---'
`o
`200 400 600 800 1000 1200 1400 1600
`Plasma AUG (nmol/L • h)
`
`•
`
`• •
`••
`~~.
`
`r2= 0.21
`P = 0.004
`
`•
`
`500
`
`0",
`
`0)
`c
`:5_
`-t:;.s= 400
`~.:.,
`.E ~ 300
`<.)3
`::)",
`«0) 200
`~"O
`",:;J
`- 0
`20)
`(i)(3
`0:>
`~ c
`E
`
`100
`
`0
`
`0
`
`800
`600
`400
`200
`Plasma AUG for cladribine x dGk activity
`(~moI/L • h x pmol/mg/min)
`
`1000
`
`Fig. S. (Top) The relationship between the area under the con(cid:173)
`centration·time curve (AUC) of plasma cladribine and intracel(cid:173)
`lular cladribine nucleotides in 69 patients with chronic
`lymphocytic leukaemia, hairy cell leukaemia and acute myeloid
`leukaemia treated with cladribine 0.085 to 0.22 mg/kg intrave(cid:173)
`nously or subcutaneously or 0.24 mg/kg orally. (Bottom) The
`relationship between intracellular cladribine nucleotides and the
`AUes of plasma cladribine multiplied by the activity of the phos(cid:173)
`phorylating enzyme deoxycytidine kinase in leukaemic cells in
`37 patients with chronic lymphocytic leukaemia, hairy cell leu(cid:173)
`kaemia and acute myeloid leukaemia treated with cladribine
`0.085 to 0.22 mg/kg intravenously or subcutaneously or 0.24
`mg/kg orally.
`
`9_ Conclusion
`
`Pharmacokinetic data on cladribine are slowly
`accumulating, and now allow more definite recom(cid:173)
`mendations for its administration. Taken together
`with the available clinical data, there is no support
`for the use of continuous intravenous infusion of
`cladribine; subcutaneous injection can supersede
`intermittent intravenous administration. Finally,
`due to its variable bioavailability, oral administra(cid:173)
`tion is recommended when repeated courses are
`given (i.e. for most indications except hairy cell
`leukaemia).
`
`References
`I. Christensen LF, Broom AD, Robins MJ, et al. Synthesis and
`biochemical activity selected 2,6-disubstituted-(2-deoxy-a(cid:173)
`and ~-D-erythro-pentofuranosyl)purines. J Med Chern 1972;
`15: 735-9
`2. Giblett ER, Anderson JE, Cohen F, et al. Adenosine-deaminase
`deficiency in two patients with severe impaired cellular im(cid:173)
`munity. Lancet 1972; II (786): 1067-8
`3. Seto S, Carrera CJ, Kubota M, et al. Mechanism of deoxy(cid:173)
`adenosine and 2-chlorodeoxyadenosine toxicity to non-dividing
`hunam lymphocytes. J Clin Invest 1985; 75: 377-83
`4. Kazimierczuk Z, Vilpo J, Seela F. Base-modified nucleosides
`related to 2-chloro-2'-deoxyadenosine. Helv Chim Acta
`1992; 75: 2289-97
`5. Beutler E. Cladribine. Lancet 1992; 340: 952-6
`6. Carson DA, Wasson DB, Beutler E. Antileukemic and immu(cid:173)
`nosuppressive activity of 2-chloro-2'-deoxyadenosine. Proc
`Natl Acad Sci USA 1984; 2232-6
`7. Carson DA, Wasson DB, Kaye J, et al. Deoxycytidine kinase(cid:173)
`mediated toxicity of deoxyadenosine analogs towards malig(cid:173)
`nant human Iymphoblasts in vitro and towards murine Ll210
`leukemia in vivo. Proc Natl Acad Sci USA 1980; 77: 6865-9
`8. Bryson H, Sorkin E. Cladribine: a review of its pharmacody(cid:173)
`namics, pharmacokinetics and therapeutic potential in the
`treatment of haematological malignancies. Drugs 1993; 46:
`872-94
`9. Santana V, Mirro J, Cherrie H, et al. Phase I clinical trial of
`2-chlorodeoxyadenosine in pediatric patients with acute leu(cid:173)
`kemia. J Clin Oncol 1991; 9: 416-22
`10. Juliusson G, Heldal D, Hippe E, et al. Subcutaneous injection
`of2-chlorodeoxyadenosine for symptomatic hairy cellieuke(cid:173)
`mia. J Clin Oncol 1995; 13: 989-95
`II. Piro LD, Elison OJ, Saven A