Bioavailability of Higher Dose Methotrexate
`Comparing Oral and Subcutaneous Administration in
`Patients with Rheumatoid Arthritis
`MONIQUE HOEKSTRA, CEES HAAGSMA, CEF.S NEEF, JOHANNES PROOST, ANTONIUS KNUIF,
`and MART van de LAAR
`ABSTRACT. Objective. To dctennine the bioavailability of higher om] do$cs of methotre.i:atc (MTX) in. adult
`patients with rheumatoid arthritis (RA).
`Methods. A. phamiacokinetic analysis was performed in 15 patient� with RA taking 11, stable do�e of
`l\ITX (� 25 mg weekly). Separated by 2 weeks, a pham1acokinetic analysis was p�rfo:rrned in each
`patient after oral and subcut&ncous administration of the same dose of MTX. MTX serum coucen·
`trations were measured by a fluoreMcnce polari?.Ation immunoassay. Pharm.acokinctic aMlysis was
`pcrfom,.ed with an iterative 2-stage Baye.si1m population procedure, obtaining population and indi·
`vidual pham:iacokim:tic parameters.
`Re:m/.ts. The median MTX dose was 30 mg weekly (range .25-40 mg). A 2-compartment model be�t
`described the serum MrX concentration versus time curves. tlie mean bioavailability after oral
`MTX was 0.64 (range 0.21-0.96) compared to lll) bcutancous administrttion. There was a st.atistically
`significant dlf'ference in the bioavailability of the 2 administration regimens.
`Conclusi,m, Bloavaila.bility of a highc:r oial dose of MtX in adult patients with RA is highly var.i·
`able .. and on average two-thirds tha.i·. of the subcutaneous administration. To improve effi.caey of
`M1X at dosages of 25 mg weekly or more, a change to pntcnteral administration should be consid·
`ered, (J Rl:teumatol 2004:31:645-8)
`Key Indexing Terms:
`METHOTREXATE
`PHARMACOKJNETICS
`RHEUM A.TOJO ARffiRJTIS
`patients with RA �how comparable bioavailabi.lity of oral
`Methotrei1:ate (MTX) i is commonly used �n weekly single­
`dose regimens in the t�atment of rheumatoid arthritis (RA).
`and parenteral MTX in doses up to 25 mg week1y8·ll. In
`A dose�effcct relaboni was establi.shed for do:.-es of 7 .5-25
`these studi.es the mean relative bjoava.ilabilit:y of oral J\,ITX,
`mg per wee]c1•2• In clinical trials in RA, the MTX close is
`compared to inttamuscu!ar administration, ranged from 0.85
`to 1.0. In other studies, using 15 mg MTX �nd 10 mg/mZ
`increased up to 25 mg!weekly, until efficacy is reached. It .is
`l\.fTX, bioavailability of oral compared to intravenous MfX
`not clear whether even higher oral doses of MTX are more
`• In a comparison of 25
`effective. Efficacy of high intravenous doses of MTX
`was 0.67 and 0.70, respectiveJy l 2
`(40-500 mg/m2), in : patients with refractory RA, wa�
`mg MTX, the me.an bioavailability after oral administration
`was 73% compared to the intravenous rou.tcH. Despite the
`described .in several studies3
`imprcssi.on given by a few studiesR. i:1, it is not certain that the
`The bioavailabiJity of oral MTX could be a limiting
`factor for its cfflcacy. Oral MTX is absorbed in the proximal
`bioava.ifability of intravenous, intramuscular, and subcuta­
`intestine by .a specific transport mechanism. and a relation
`neous MTX is strictly comparable.
`between dose and absorption of oral MrX was observed in
`Phannacokinetic studies in patients with maJignant
`diseases have shown that the absorpti.on of higher doses of
`2 clinical stuclies6.7
`• · Pharmacokinetic studi.es in adult
`MTX (� 25 mg weekly) is incomplcte6• 1 6- 19• The relative
`hioavaibhility of 40 mg/m2 oral .MTX in a study in childnm
`with acute lym.phoblastic leukemia was 42%; in adult
`patients with solid tumors using 15 mg/m2 MTX this was
`57% 1 R, l7. Another study in 1 5 children 19 showed a decreased
`absorption of oral MTX at doses > 12 mg/m2• However, the
`results of pbarmacokinetic studies in disorders other than
`RA, and even more so in children, cannot be extrapola.ted to
`adult patients with RA.
`Although a dear rclation between ph&rmacokincti.c para�
`mctcn. and efficacy has not been demonstrated in RA, i.t
`seem� likely that improvement of the bioavailabi.lity of
`MTX will lead to better efficacy, given the dose-effect rela-
`
`1
`
`-
`
`.
`
`F'!'r;n, tJ,e J),1partmf.t1i ofl,.iuitlmu.wl,;gy, ,md t/ie Deparl�enc o[Clinical
`Pharmacology. Medisch Spcttrwn 1lvente, enschede; and th!!! .Ditpartm�nt
`uf Pharmacaki�tics artd Drlig Delivery, U11iv1trsity Centre fer Pharnuicy.
`Grcmingen, The Netht!rlands.
`M. llr,,ekstra, MD. Rltcu11U1tol()gis1; C.J. Haag� MD. Rhtrunatolcgist;
`M.IJ..FJ. van de Laa,: MD. Rhe1,m.ato/()gis1, Department ()f Rheum(}./Olcigy.
`Ml!di.,dr :Spccrrum 7wcnte; C, Neef. PhD, Ho.,pital Pha,mtzt:ist; A..T,H.
`&r,if. f.>harmaceutlcal Analys1, Depal'fment of Clinical Pharmar.ology,
`Mtdisch Spectrum 1wente: J.H. Pror>sl, PhD, Pharm,:u:isr, De.partmenr of
`1'harmacokinetics and Drr;g D11/ivNy, llnivcrsiry Centl"t for P�rmacy.
`A.d4re.ts rttprittt n!que.m to Dr. M. Hodmm, Departmenr of
`khtwtrlll(I/Oty, Mtdlsch Sper.ll't.lm Twenlc, PO BM< $0000, 75()() KA
`l!�diedt!, 'The Netherlemds. E·tttQi/: 111.hookstra�ii,.,J,;erifl.u/.Ml'l.,t,nf
`Sllbmitt<!d May 22, 2()()J; revision ar.r.�prcd Or.tobtr 17. 2()()J,
`
`I
`
`·
`
`J ;
`,
`1 ; ,
`
`·~
`
`1·1 :\
`'·f
`
`1 3
`
`,
`
`645
`
`Page 1 of 3
`
`KOIOS Exhibit 1004
`
`

`
`'i
`
`tion 1.i. This idea is supported by a sCudy in patients with
`psoria sis in which a relation was found bet.ween the area
`under the curve of the time versus MTX concentration and
`in the Psoriasis Activity
`:ind Severity Index
`a decrease
`(PASI)2o_
`of higher MTX dol;es can be
`The bioavaHability
`improved by parenteral administration_ To study this option ,
`study in adult
`we performed ~ crossover phaonacok.itietk
`patients with RA, comparing the bioavaUability of oral and
`subcutaneous MTX at doses 2: 25 mg weekly ..
`
`MATERIALS AND METHODS
`Patif!nts arrd MTX admi11istrati011, P11tlcnt~ with RA. who were ~ated with
`Ml'X in a ,table ~ 3 months) dMe of~ 25 mg wecldy, oml or p~rcntcrt1!.
`were studied. Consccoti1rc ou!f!ati11.nt~ fulfilling thcl!C inclltsron criteri;t
`Wett: invited to participate, The local ethiCR Commj~e a.pptoved the Study
`and written informed eon~ent was obtained fr.om C,QCh patient.
`Baseline da~ were $athercd on diagnosui, age. sex. disca.se du-ration.
`dose , sCl\lm crestinine, fulie acid supp{emen1;1tion, Md use o( diRea~e
`rnodi.fyintl antirhc:umatic drngi. (DMARD) , ttollstetoidal antiinllammatory
`dn.tgs {NSAID), and prednisol~. Pham1scoldnetles were studied twice in
`each paticl\t with a 2-ll'eck intcrvsl: om:c with Choir regulor MTX do~ b)'
`01111 r.oure of administnitioo. and O!JCC with the ssme dose of M'tX by
`N1'bcu!Jllle0u~ admimsmtioa ic random Qrdcr. Folic acid supplement.t.tion
`w~~ allowed, but not on the day of MTX it\talce. leukopeni.a, cbrornbo(cid:173)
`for eJtclURiOn.
`cytopcnia, and transamina~ elevations were ~sous
`Patien~ wen: adll:littc4 i11 the hospit.'11 in tbc morning. They were
`,1llowcd to have breald'est at lt<J1t11:, at least l . .S hour Wore MTX lhtllkc.
`. was continued dunng both sampling epiRodes. Other
`Cotn~icatton
`DMA RD and p~¢dni~c,ne \l!el'C allowed. with stable doses tl\J'oughout the
`t..'1<:en at leaiit 1.S hour before and
`~tUdy, The concum:11t medication W(ls
`more t.lllln 2 b01its ~r M1'X tntalcc. Oml M'JX wa~ adminlsrercd wi.th
`watc~. MTX was injected subcutaneous}}' in th.e UPJ)el' le$ .in :ill pntient~ by
`the examiner. Blood ssrnpl.e$ were drawn from .an indwellio,g esthece. at
`Tlrne O (preadroiniwatlcrn) and at 0,25, 0.5, O. 75, 1.0, L2S, I .S, 2, 4. 6. 8,
`l2. 24, and 48 h a'1er administration of MTX. The blO<XI ~mple~ w-cre
`centrifuged and the scrum !Jll)fl::d at -20°C until an~ly~is.
`M7X atsay. MTX 1<erum concentrations were determined ll~ing a fli.ore!(cid:173)
`ccnce polarization immono.:is.~sy rechruqllc (MTXII ; Jlst no. 7A12, TOX·
`Abbott Diagnosli:i;s, North Chicago. n.. USA)2'. The lower detection limit
`was 10 µg/1. At JO µg/1 the cocf:ficlent of vadn4on of the tci:t is 15%. The
`iitanoard devistion (SD) of the tC5t iR described by the fonnuls : SD "'"4.76
`+ 0,0S•C , WIJerc C "' COll<lenttlltiOn.
`('l,arniocokineric aru:t.lyJis. The MTX COfleentration dat.., of both adm.ini.s(cid:173)
`tmti<JM from all p0tients wen: am1l~d simulcanoously by fln iterative 2.
`stage Bayesian :vialysis using the program MW\Pherm, v~icm 3.54l2.2' .
`Thi! phannacokiuetic model was a one~partment
`(i,arametffl k., V 1) or
`It,., Vt• k1~. ~ 1), with first-order
`• 2~ornpartment model (p~
`sbsoq,tion with a llll:·time for (lfll{ t>'"' ~ub(:~1~ncc-~, ad,niiti~~~tfon. wltll
`inunroeters F (b!011vailability), k, (absorption ni.11, constant). ind T !of (lag(cid:173)
`time) for each route -0f administratlo11. Sil\Ce. absolute l)ionvAllability cannot
`be a:1se~sed wi !bout rui intr:\vcnous reference administr.atior1, tM analysis
`w~s perronoed .assuming th11t bioovailsbility of tltc subc:utancou~ adminis(cid:173)
`~on wBS 100%. Mea~url!ment datll were weighted nccerding to the reci·
`(USD2). A \og,nom,aJ dlstr.ibution fot the
`proc11l Qf their v~ance
`ph.vmacokinetic popuhltion panunctcn W-\S assumct"l. Goodncss-<>f-fit was
`evaluated from vbual inspection of the measured and e:!.lcule.ted dsta
`poinu. The choice between ~ 01JC-.,nd 2-compartment model wa.~ ba...<'.Cd on
`Alcaike'~ Ioformation <:riterion (AJC)'A·
`MTX clcanm~ (CL). volutne of dillttibotion (V ) . climlriation half-life
`(t~ ). a~ for e11ch roure of odminismrtiOII the~ Ulldcr the concentn11i0r1-time
`J7f'(lfilc (A.UC), tif1"1(: to m.l:!lirnwn conceall'atioll (T .... }. and ITl4Ximum concen-
`
`tmtion (C ..... ) were e<1IC\1l~tcd l'rom 1hc model narnme(cr.i: (or c~c!i patic111,
`
`Stari.vtir:al ana l,v.tf,,·, T~, compnrc 1hc v~luc., of t11c pharm:1cOkin<itic p:ir:illl(.
`tel':l nfthe (\flll and ~ubcu1:incuu.< rnut11 nfn<lmini.~tration. s si~!'lcd-r~nk ~
`wu~ cmployc,:t. /\ 2·~idcd p v11h1c-;; 0.05 was considered ~ignilkan t,
`
`RESlJL'fS
`Fifteen patients with RA were :;;tudied. Patient char.acteti~
`tics are presented in Table l . All patients received folic acid
`supplementation
`in varyi.ng doses (5- 25 mg weekly), but
`not on the day of MTX intake . Three patients concurrentJy
`used hydroxychloroquine, one chloroquine, one sulfasa.
`lazine, and one aurothiomalatc. Low dose prednisolone
`(S 10 mg daily) was used by 8 patients, and NSAJD by 11
`patients.
`A 2-compartment model fitted significantly better to the
`data than a one~compartment model (A1C value -250 and
`-956 for the one · and 2-compartment model. respectively).
`The mean bioa\rai1ability (F) was 0.64, w.ith a rather large
`range from 0.21 to 0 .96. The pharmacokinctic parameters
`with pa.ired statistical analysis are shown in Table 2. The
`AUC of oral MTX was significantly lower than the AUC of
`the subcutaneous route of administration (p < 0.001). The
`fitted mean ti.me-concentration curves of oral and subcuta(cid:173)
`neous adrninistr.ati.on are presentccJ. in Figure 1.
`
`DISCUSSION
`The bioavadability of oral MTX (~ 25 mg weekly) was
`highly variable, and was significantly
`less compared
`to
`subcutaneously administered MTX in patients with RA. It
`varied between 0.21. and 0.96, with a moan of0.64.
`In the design of our study comedication was continued,
`and patients were al.lowed to have breakfast at home before
`comfag to the hospital . Because of the time between comcd(cid:173)
`ication, breakfast, and MTX administration, an effect on
`MTX absorption is unlikely. Furth.er, the effect of food ba.c;
`been extensively studied and no effect on MTX absorption
`was found 9•13 .
`The majority of pharmacokinctic studies in adult patients
`with RA have used low doses of MlX. In studies using
`MTX doses of7.5 to 20 mg weekly. bioavailability after oral
`compared to parenteral administration ranged from 0.67 to
`1.QR·t;. Only one study cotnpa.n:d 25 mg oral and intra·
`venous MTX, in 18 patients wi.th rheuir.:1tic diseases 11. TI1c
`tJioavailability of oral Ml'X was 0.73, somewhat higher th.an
`what we found, but in our study mo~t patiCJJts used higher
`
`Table 1. Patient char.1eteri!ltic$ (n.,., 15: 11 women. 4 men).
`
`Age. yrt.
`DiSc.1se duration. YJ1l
`Weight, lc;g
`Crc:atininc dcsttu1cc , ml/mit1
`Dose. mg weekly
`Do~e. mg/l<;g
`
`Mediao
`
`~imgc
`
`61
`7
`76
`80
`30
`0,40
`
`ll-n
`2-32
`63-110
`57-124
`2.5--40
`0 .27---0.57
`
`The Journal of RhcumatQlo,i:y 2()()4; 31 :4
`
`Page 2 of 3
`
`KOIOS Exhibit 1004
`
`

`
`Ta,·M 2. Pharrnacokinctic par,1mcter~ of t)ral ,ind s11bcut:1ncou~ roulc t•f :id min isl rntion (n : 15 ). Signcd·rank. les t , p vuluc "' 0.05 i.s signifi<;@ I:.
`
`Ot~l
`
`Svhc.utaneou.s
`
`p
`
`AUC
`
`Lag·timc
`
`ks
`
`T,,,.ax
`
`cmll~
`
`2466
`(785)
`3786
`(873)
`0.001
`
`0.36
`(0.18)
`0.06
`(0.05.l
`<0.001
`
`0.87
`(0.29)
`0.36
`(0. TO)
`< 0.001
`
`1.2
`(0.3)
`1.7
`(0.3)
`0 .001
`
`594
`(20!l)
`519
`(142)
`0 .30
`
`VI
`
`9 .(,
`(2.0)
`
`v
`
`34.:'i
`(8. l)
`
`kl':?
`
`k21
`
`ke.
`
`CL
`
`t 1h cl
`
`ORl
`(0,31)
`
`0.55
`(0 .0·l)
`
`0,88
`(0.l l )
`
`8 ,4
`(2.2)
`
`2.9
`(0.5)
`
`Values arc mean (sumdard deviation). AUC: area uQde:r curve: (0-48 hours) in J1·µW}; l&R·timc in hour.s; ka: abR()f!)tion rate con~tant; Tm.,: time to ma,i:imum
`c:onccntration (hours); Cm.,: ma.,;i.mum C(Jacentr.1tion (µg/1); Vl: volume of dintribution of fust comp:u:tmcnt; V: \rQlumc of distribution ()lter); !<12: ~ate
`constant of Cransport between compartment L and 2; k21: rate c-onmam of transport between cmnpartrncnt 2 and l; ke: diminatio~ mte coMtant: CL: total
`l,ody clearance (liter/hour), t 1h. el: halt'-1.ife of elimination (h(lun.).
`
`fJ(J(l
`
`soa
`
`.5
`
`f 400
`·~
`c 1300
`..,
`§
`" 200
`Si
`
`100
`
`0
`
`0
`
`4
`
`a
`
`16
`
`20
`
`24
`
`1:?
`time In hou~
`I'iP,utF l . Plasma concentration-time curves of oral (or) and ~ubcutaneou•
`(sc) methotteutc, Values arc means,
`
`doses than 25 mg.
`When we compare our data to other studies using higher
`do:se MTX (> 25 mg), only pharmacokinctic studies in
`patients with malignancies are available. In these studies
`in MTX absorption was obscnred, and
`wide variability
`therefore split-dose regimens have been tried to impro\re
`bioavailahility25. A comparable lnvestig~tit-n is the study by
`Freeman-Narrod, et al. Doses of 15 mg/m2 (25-35 mg) were
`used in adults with solid tumors. Eighteen patients received
`this dose by both oral and intramuscular administration. The
`mean cu.mutative AUC up to 24 h wa$ higher with the intra(cid:173)
`muscular route, and the mean oral bioavailability wail
`0.5717•
`'
`We analy,..ed the data assuming flr!;t-order absorption
`after a lag-time, which may be a simplification of the true
`absorption kinetics. tn general a difference in AUC betweeo
`or.al and subcutaneous administration of medication could
`be due to either an absorption limitation or a first~pass
`
`effect. Decreasing bioavatlability with an increasing dose
`favors an absorption limitation. The number of patients in
`our study with different .MTX doses was too small to draw
`conclus.ions about a dose-bioavailability relati.on. However,
`there is a positive relation between the subcutaneous AUC
`and the dose of MTX (Jin.ear regression; R2 "" 0.33, p ;:;:
`0.03), whereas the o.raI AUC does not increase with an
`increasing dose. Hamilton, et al studied 21 RA patients on
`more occasions. They found a decreasing bioavailability
`with an. increasing oral dose, mean maximum dose being 17
`ro.g/week7. These results support the idea of an absorption
`limitation of oral MTX with an increasing dose. The finding
`of higher bioavailability of oral split high dose J\.ITX:,
`compared to a single dose, in patients with solid tumors25
`supports a reduced bioavaUabiJity due to an absorption limi(cid:173)
`t.a.ti.on. However, to pursue thi.s question for the MTX doses
`we u~e in RA, an additional. study is needed that directly
`compares a single-dose with a split,dose regimen.
`Although controlled trials studyi.ng the effect of higher
`doses of MTX are lacking, higher dosing of MTX may be
`clinically useful. A dose escalation study in 54 patients with
`RA concluded that increasing the intramuscular MT.X dose
`from t5 to 45 mg weekly did not result in improved disease
`control26• However, the number of patients was small, and
`detailed data about baseline disease activity scores were not
`provided. In our opinion, additional controJJed trials are
`needed to evaluate the effect of higher doses of MT.X. which
`are in fact widely used in rheumatology practice. In our
`obsi::rvational study of l..:ITX use in 1022 RA patieuts, we
`found that 12% of the patients reached a rnaximum dose of
`~ 25 mt weekly (maximum 40 mg weekly)27.
`Our data suggest that doses between 25 and 40 mg MTX
`per week, administered ora.Uy, result in limited bioavail(cid:173)
`abi1lty. Bioavailability is enhanced by the subcutaneous
`route of administration, and this ma:,, increase efficacy.
`
`REFERENCES
`J. Purst DB, Koehn kc R, Burmci$tet LF.', Kohler J, C(ugi II r.
`Ittcroasin.l!'. mothotrcxate effect with lncrca~ing do~e in the treatment
`of J'eAiAtant rt,eumatoid arthriti~. J R'1eumatol 1989;l6:313-20.
`the rc111iionRl1iJ) bc(ween do~e and
`'-· Seidcmo.n I>. M<>thnh'e""t" -
`
`.,,
`
`' ,,
`··'I ,.,
`
`•
`
`No,:lr,strrz, et aJ: Biottvai/4bilil:,, of M.1'X.
`
`647
`
`I
`
`Page 3 of 3
`
`KOIOS Exhibit 1004