Ciinicai Cancer
`Research
`
`Clinical pharmacokinetics of the antipurine antifolate (6R)-5,10-
`dideaza-5,6,7,8-tetrahydrofolic acid (Lometrexol) administered
`with an oral folic acid supplement.
`
`S R Wedge, S Laohavinij, G A Taylor, et al.
`
`Clin Cancer Res 1995;1:1479-1486. Published online December 1, 1995.
`
`Updated Version
`
`Access_the most recent _version of this article at:
`http://cl:ncancerres.aacrjoumaIs.org/content/1/12/1479
`
`Citing Articles
`
`This article has been cited by 3 Highwire-hosted articles. Access the articles at:
`http://clincancerres.aacrjouma|s.orglcontent/1/12/1479#re|ated-urls
`
`E-mail alerts
`
`Sign up to receive free email-alerts related to this article or journal.
`
`Reprints and
`subscriptions
`
`To order reprints of this article or to subscribe to the journal, contact the AACR Publications
`Department at pubs@aacr.org.
`
`Permissions
`
`To request permission to re-use all or part of this article, contact the AACR Publications
`Department at permissicns@aacr.org.
`
`Downloaded from clincancerres.aacrjouma|s.org on February 1 , 2011
`Copyright © 1995 American Association for Cancer Research
`
`Lilly Ex. 2048
`Sandoz v. Lilly IPR2016-00318
`
`

`
`Vol. I. I-I79-I486. Dt'(‘t‘Illht’!' I995
`
`Clinical Cancer Research I479
`
`Clinical Pharmacokinetics of the Antipurine Antifolate (6R)-5,l0-
`
`Dideaza-5,6,7,8-tetrahydrofolic Acid (Lometrexol) Administered
`with an Oral Folic Acid Supplement‘
`
`Stephen R. Wedge, Sudsawat Laohavinij,
`Gordon A. Taylor, Alan Boddy,
`A. Hilary Calvert, and David R. Newell’
`Cancer Research Unit. The Medical School. University ol' Newcastle-
`upon-Tync. Framlington Place. Newcastle-upon-Tyne. NE2 -'tHH.
`United Kingdom
`
`ABSTRACT
`
`(GR)-5.10-Dideaza-5.6.7.8-tetrahydrofolic acid (iome-
`trexol) is an antipurine antlfolate which selectively inhibits
`glyclnamide rlbonueleotlde formyltransferase. Lometrexol
`pharmacokinetics were evaluated in l7 patients (32 courses)
`as part of a Phase I study in which folic acid supplementa-
`tion was used to improve tolerance to the drug. its clinical
`utility being previously limited by severe cumulative toxic-
`ity. Lornetrexol was administered as an i.v. bolus every 4
`weeks at a starting dose of 12 mglm’. with subsequent
`interpatient dose escalation to 16. 30, and 45 mg/m’. p.o.
`folic acid (5 mglday) was given for 7 days before and 7 days
`alter lometrexol administration. The disposition of total
`lometrexol in plasma was best described by a hlexponential
`model for data acquired up to 12 h after drug administra-
`tion, although triexponential plasma phannacoltineties were
`often found to give a more adequate description when data
`were available at later time intervals (24 h and greater).
`Mean plasma half-lives (: SD) for model-dependent analy-
`sis were (met I9 3: 7 min, (":3 256 1 96 min. and (my
`(where measurable) I170 1: 435 min. Lometrexol area under
`plasma concentration versus time curve was proportional to
`the dose administered. Moderate plasma protein binding of
`lometrexol was evident (78 :1: 3%) with an inverse linear
`relationship between fraction of unbound lometrexol and
`the concentration of serum albumin. The volume of distri-
`
`bution of lometrexol at steady state was between 4.7 and 15.8
`l/tn’. Renal elimination of lometrexol, studied in l9 patients
`(21 courses), was considerable. accounting for 56 1: 17% of
`the total dose administered within 6 h of treatment. and 85
`:l:
`l6% within 24 h of treatment. These recoveries of un-
`changed Iometrexol indicate that the drug does not appear
`to undergo appreciable systemic metabolism at the range of
`concentrations studied.
`
`Lometrexol pharmaeokinetlcs were also examined in
`seven patients who received 45 or 60 mg/m’ lometrexol as
`part of a separate study of the drug given with folinic acid
`rescue 5-7 days after treatment. No marlted differences
`were evident in lometrexol plasma half-lives. plasma clear-
`ance. or the extent of plasma protein binding. indicating that
`there is not a pronounced pharrnacokinetie interaction be-
`tween lometrexol and folic acid.
`
`INTRODUCTION
`lnmetrexnl is a fnlnte analogue which selectively inhibits
`GAR“ fonnyltransferztse. an enzyme essential for dc nova purine
`biosynthesis (l. 2). This antipurine antifolate exhibits a broad
`spectntm of antitumor activity in marine and human xenograft
`tumor models. in which the established antifolate methotrexate
`demonstrates little or no effect (3).
`in early Phase iciinicai studies with iometrexoi. significant
`clinical toxicity was evident. characterized by severe mucositis
`and myelotoxicity (thrombocytopenia and leucopenia). which
`limited drug administration to only one or two courses (4-7).
`These toxicities were unexpected. occurring at drug concentra-
`tions which were approximately one hundredth of the l0%
`lethal dose in mice (8). A number of clinical responses were
`documented. including activity against malignant fibrous histi-
`ocytoma (5). non-small cell
`lung cancer. breast cancer. and
`colonic adenocarcinoma (7). which stimulated studies aimed at
`the pharmacological amelioration of lometrexol toxicity. Exper-
`imcntation in mice revealed that the therapeutic index of lame-
`trexol was highly dependent on dietary folic acid intake (9. l0).
`and suggested that folic acid administration could reduce tox-
`icity. without ablating antitumor activity. To enable the devel-
`opment of a tolerable and effective schedule for the routine
`clinical use of lometrexol. a Phase l study was initiated in which
`5 mg folic acid/day were given for 7 days before and after
`lometrexol administration. After 7 days. this dose of folic acid
`resulted in an increase in plasma folate levels from 3 to 64 ng/ml
`to 6 to I80 ng/ml in the patients studied.
`The Phase I study of lometrexol with folatc supplementa-
`tion provided an opportunity. for the first
`time.
`to conduct
`detailed clinical pharmacokinctic studies with lomctrcxol: a
`comprehensive pharmacological examination in humans being
`previously prohibited by the lack of clinical utility and avail-
`ability of a suitable assay. The principal objectives of this
`
`Received 2/24/95: revised 8Il5/95: accepted 8ll7I95.
`' S. R. W. and S. L. were supponed by Eli Lilly and Company (India-
`napolis.
`lN). Financial support was also provided by the North of
`England Cancer Research Campaign.
`3 To whom requests for reprints should be addressed. Phone: -l4-l9l-
`222-8233: Fax: 4-I-l9l-222-7556.
`
`"The abbreviations used are: GAR. glycinamide ribonucleotide: CV.
`coefficient of variation: AUC. area under the lomeuexol plasma con-
`centration r-er.ru.r time curve: Cl.,m-. total plasma clearance: GFR. glo-
`merular filtration rate: Vd“. volume of distribution at steady state: FBI’.
`folatc-binding protein.
`
`Downloaded from clincanoerres.aacrjouma|s.org on February 1, 2011
`Copyright © 1995 American Association for Cancer Research
`
`Lilly Ex. 2048
`Sandoz v. Lilly IPR2016-00318
`
`

`
`1480 Lometrexol with Folic Acid Supplementation
`
`(a) to determine the
`clinical phannacokinetic study were:
`plasma pharmacokinetics in patients receiving multiple courses
`of lometrexol. and thereby define the relationship between
`lometrexol AUC and dose. and intra- and interpatient variability
`in AUC; (b) to investigate the extent of lometrexol plasma
`protein binding; (c) to measure urinary excretion of lometrexol;
`and (d) to evaluate the effect of folic acid supplementation on
`lometrexol pharmacokinetics to determine whether the im-
`proved tolerance of lometrexol produced by folic acid adminis-
`tration is a consequence of a pharrnacokinetic interaction.
`
`MATERIALS AND METHODS
`Patient Eligibility. Patients eligible for this study had a
`histiologically documented malignant solid tumor. which was
`either refractory to established therapies. or for which no stan-
`dard therapy existed. All patients had a predicted life expectancy
`of at least I2 weeks. and had recovered from the toxic effects of
`previous treatment before entering into the study. i.e.. they had
`not received any major therapy or irivestigatiorial drug for at
`least 4 weeks (6 weeks if prior therapy included chemotherapy
`with a compound known to have delayed toxicity, e.g.. a nitro-
`sourea). Exclusion criteria included factors which could have
`interfered with lometrexol disposition/toxicity or folic acid absorp-
`tion and corrprised: (a) concomitant medication with probenecid.
`trirnethoprim. co-trimoxazole. pyrimetharnine, prednisolone. anti-
`epileptics. or allopurinol; (b) extensive radiotherapy; and (c) in-
`flammatory, ulcerafive bowel disease or malabsorption syndrome.
`All patients were required to have adequate organ function
`prior to ueatment. with hepatic function characterized by bil-
`irubin levels of <25 umol/liter. and renal function by a creat-
`inine measurement of < I20 p.mol/liter and a “Cr-EDTA clear-
`ance of >50 mllmin.
`lnfonned written consent was given
`according to local regulatory requirements.
`Study Design. Folic acid (Approved Prescription Ser-
`vices Ltd.. Leeds. United Kingdom) was given daily as a single
`5-mg tablet for 7 days before and 7 days after lometrexol
`administration at 4-week intervals. Lornetrexol (Lilly Research
`Centre. Surrey, United Kingdom) was reconstituted in 0.9%
`(v/v) saline and administered as a rapid i.v. bolus over 0.5-1.0
`min at a concentration of I-I0 mg/ml. Patients were admitted to
`the Department of Medical Oncology. Newcastle General Hos-
`pital, to receive lometrexol and were observed for an additional
`24 h after drug administration to ensure that acute toxicity was
`not apparent. The perfomiance status of patients was assessed at
`least once a week. for a period of 4 weeks, following lometrexol
`therapy.
`The trial design required three patients. previously un-
`treated with lometrexol. to be treated at each dose level. The
`first patient entered at each dose level was followed up for 3
`weeks before the next patient was entered. At least two patients
`per dose level received two courses before dose escalation.
`Toxicities were evaluated according to the WHO criteria. If
`repeated courses at a given dose level were tolerated without
`toxicity greater than WHO grade 11. doses were escalated ac-
`cording to the clinical judgment of the investigator with ap-
`proval of the Medicines Control Agency (London. United King-
`dom) and the Local Ethics Committee. The starting dose of
`lometrexol was l2 mg/m’. with subsequent escalation to l6, 30.
`
`and 45 mg/m3. Dose escalation increments were determined by
`clinical experience at the previous dose level and by data from
`a parallel study of lometrexol given with folinic acid (I I). No
`inu'apatient dose escalation occuned.
`Pharrnacokinetic Studies. Lometrexol pharmacokinet-
`ics was determined in l7 patients (32 courses) receiving folic
`acid supplementation and in an additional 7 patients (7 courses)
`who did not receive folic acid. Plasma samples from patients
`receiving lometrexol without
`folate supplementation were
`kindly provided by Drs. C. Sessa and F. Cavalli (Ospedale San
`Giovanni. Bellinzona. Switzerland). who were responsible for
`an alternative Phase I study that involved folinic acid adminis-
`tration (l5 mg every 6 h for I2 doses). starting 5-7 days after
`treatment with lometrexol (l l).
`Blood samples were collected by venipuncture into vacu-
`tainer tubes placed on ice and containing the sodium salt of
`EDTA as an anticoagulant. and were taken before treatment and
`at 5. I5. 30. and 45 min and at I. l.5. 2. 4. 6. 8. 12. and 24 h.
`and in some patients at 48. 72. and 96 h. after lometrexol
`administration. Samples were immediately centrifuged ( l()()() X
`g. 8 min. 4°C). and plasma was removed by aspiration with a
`Pasteur pipette. Plasma was stored at -20°C prior to analysis.
`The plasma lometrexol concentration was measured by the
`HPLC method of Wedge er al. (12). which uses derivitization
`and fluorescence detection. Briefly. patient
`samples were
`thawed at room temperature and diluted to l ml with control
`human plasma (Red Cross Transfusion Service. Newcastle-
`upon-Tyne. United Kingdom) to contain l0-250 ng/ml lome-
`trexol. Samples were further diluted (lzl) with aqueous fonnic
`acid ll% (v/V): pH 3.7] containing I00 ng C'"-desmethylene
`lometrexol (Lilly Research Centre) as an internal standard.
`Following rotary mixing and centrifugation. each sample was
`subjected to solid-phase extraction using a C8 (l cm) Bondclut
`cartridge (Analytichem lntemational. Harbour City. CA). Eluted
`samples were evaporated to dryness using a Speedvac concen-
`trator (Savant Ltd.. Farmingdale. New York) and reconstituted
`in l3% (v/v) aqueous fonnic acid. Oxidation of lometrexol and
`the internal standard was achieved by incubation (37°C. 90 min)
`with a suspension of manganese dioxide (0.2 mglml) in water
`and tenninated by the addition of a l:I mixture of 5 M NaOH
`and 1% (wlv) ammonium carbonate (pH 5) to samples on ice.
`Samples were centrifuged (l3,000 X g, l0 min). and l00 pl of
`the supernatant were analyzed chromatographically. Chromato-
`graphic analysis was achieved using an Apex ll (C l8. 3 pm: I50
`X 4.6 mm) analytical column (Jones Chromatography. Hen-
`goed. Glamorgan. South Wales. United Kingdom) and a mobile
`phase of l2% (wlv) acetonitrile in l% (v/v) aqueous acetic acid
`(pH 5) containing tetramethylammonium hydrogen sulfate
`(0.I7l g/liter) as an ion pair reagent. Elution was isocratic. at a
`flow rate of l ml/min. and analyte measurement was by fluo-
`rescence detection (Ex. 325 nm: E“. 450 nm).
`To assess intraassay variation. each assay was calibrated
`using a five-point standard curve of duplicate lometrexol stan-
`dards in the range 10-250 ng/ml. prepared in control human
`plasma. and extracted/analyzed at
`the same time as patient
`samples. Quantitation was achieved using internal standardiza-
`tion by a comparison of peak height ratios. with peak heights
`being quantified using Minichrom Software (VG Data Systems
`Ltd.. Altrincham. Cheshire. United Kingdom). All calibration
`
`Downloaded from clincanoerres.aacrjouma|s.org on February 1, 2011
`Copyright © 1995 American Association for Cancer Research
`
`Lilly Ex. 2048
`Sandoz v. Lilly IPR2016-00318
`
`

`
`Clinical Cancer Research l48l
`
`Table I
`
`Patient characteristics
`
`l 7 t_ 32)
`
`6
`I I
`
`Evaluable patients (courses)
`Sex
`Male
`Female
`Median age. yr (range)
`WHO performance status
`
`0I2
`
`Primary tumor
`Breast
`Ovarian
`Malignant melanoma
`Others
`Prior treatment
`Chemotherapy
`Radiotherapy
`
`were stored at -20°C prior to analysis. Samples were thawed at
`room temperature and diluted in control urine to ensure that the
`lometrexol concentration would be between 2 and 25 p.g/ml
`(i.e.. the dilution factor used was estimated according to the dose
`of lometrexol administered and the volume of urine produced in
`a 6-h period). The method for sample preparation was as de-
`scribed for plasma samples. except C"’-desmethylene lome-
`trexol was added at a concentration of IS irglml. and evaporated
`samples were not oxidized. but resuspended in ISO ul of the
`mobile phase and 50 it] were analyzed chromatographically.
`Chromatographic analysis ( I2) involved a Spherisorb C6 (5 pm:
`150 x‘ d.6 mm) analytical column (Jones Chromatography) and
`isocratic elution with a mobile phase of 13% (w/v) acetonitrile
`in aqueous phosphoric acid [I% (vlv). pH 3.2] at a flow rate of
`1.5 mUmin. Each assay included a five-point standard curve
`(duplicate samples within the range 0.2-l0 p.glml). and quality
`assurance standards at 0.2. 2. and I0 uglml were assayed in
`triplicate. The intraassay and interassay CV5 for these assays
`were found to be <2% and <3%. respectively. Lometrexol
`excretion in urine was expressed as a percentage of the admin-
`istered dose. and renal clearance was calculated as the ratio of
`the total amount excreted and the AUC.
`Statistical Methods. All values expresad with a margin
`of error represent the mean : SD. Levels of significance were
`calculated using Student's t test. where P < 0.05 was considered
`indicative of a significant difference between groups. The rela-
`tionship between lometrexol dose and pharmacokinetic param-
`etcrs was assessed by linear regression analysis and/or a Spear-
`man rank correlation.
`
`RESULTS
`Phamracoltinetics of bometrexol Administered with a
`Folic Acid Supplement. The characteristics of the patients
`studied are shown in Table l. and plasma lometrexol concen-
`tration—time profiles from representative patients receiving l2.
`I6. 30. or 45 mg/m2 lometrexol are shown in Fig.
`I. Data
`collected within the first I2 h of lometrexol administration were
`
`found to be best described by a biexponential eqration when
`evaluated by compartmental analysis. However. at doses of 30
`and 45 mg/m3 lometrexol and where data were available at later
`
`curves were linear (r3 > 0.997). the lower limit of detennination
`of the assay was I0 ng/ml. and the intraassay CV at nominal
`concentrations of l0. 50. and 250 nglml was always <8.5%. in
`addition. three lometrexol quality assurance samples (I0, 50.
`and 250 ng/ml). prepared every 2-3 months in bulk. were
`assayed in triplicate to assess the interassay CV. which was
`always <6%.
`Pltarmacoltinetic Parameters. Plasma lometrexol data
`were analyzed using both model-independent and rnodel-depen-
`dent analyses. In model-independent analyses. the AUC was
`calculated using the log trapezoidal rttle (I3) with extrapolation
`to infinity. using the terminal phase rate constant calculated by
`the companmental analysis. For model-dependent analyses. a
`biexponential or triexponentiai equation was fitted to the con-
`centration versus time data using a nonlinear-weighted least-
`squares parameter estimation program (ADAPT ll. kindly pro-
`vided by Dr. S. D. Z. D'Argenio and A. Schumitzky.
`Biomedical Simulations Resource. Los Angeles. CA). Data
`were weighted as the reciprocal of the estimated variance. where
`the SD of the output was assumed to be proportional to the
`estimated concentration (constant CV). The model providing the
`best fit to each data set was determined using the precision of
`the parameter estimates and consideration of the Akaike infor-
`mation criterion (14). The parameters derived were used to
`calculate model-independent (AUC. Cl-“,1. and Val“) or model-
`dependent (AUC. Cl-rm. (ma. r.,,B. and !,,2'y) pharrnacokinetic
`parameters (l3. l5).
`Plasma Protein Binding. Lometrexol plasma protein
`binding was determined using ["C]lometrexol (specific activ-
`ity, l3 p.Ci/mg). which was kindly provided by Dr. M. D‘lncalci
`(lstituto Mario Negri. Milano. Italy). This compound was radio-
`labeled at the carbonyl group of the benzyl moiety. and had a
`radiochemical purity of >88%. as determined by HPLC. Each
`patient plasma sample (I ml) was spiked with 3.5 pg l"'C]lome-
`trexol. rotary mixed. and an aliquot (100 pl) was removed. The
`remainder was then subjected to ultrafiltration using an Amicon
`Centrifree Micropartition Unit (Amicon Corp.. Upper Mill.
`Stonehouse, Gloucestershire. United Kingdom) and centrifuga-
`tion (1000 X g. l0 min, 4°C). after which an aliquot (I00 id) of
`ultrafiltrate was removed. ["’C]bometrexol in the ultrafiltrate
`and prefiltered plasma was determined by liquid scintillation
`counting. and the ratio was used to caicuiate the percentage of
`unbound lometrexol. Nonspecific binding of ["‘C|lometrexol to
`the filter was determined to be <4% by ultrafiltering solutions
`of ["C|lometrexol in PBS (0.l M. pH 7.4;
`1 ml) and was
`therefore ignored. Samples of plasma from each patient. taken at
`<| h and at
`l2—l20 h after lometrexol administration were
`examined to assess any potential concentration dependency of
`lometrexol plasma protein binding. but no significant differ-
`ences in binding were found (P = 0.l l. paired I-test). The mean
`of each pair of analyses was therefore used to: (a) relate the
`unbound fraction of lometrexol with serum albumin and plasma
`protein concentrations and (b) calculate the unbound Vd” (I5).
`Urinary Excretion. Urinary excretion was studied in 13
`patients (I4 courses) for which there was evaluable plasma
`pharmacokinetics available. and in an additional 6 patients (7
`courses) for which plasma pharmacokinetics was not measured.
`Urine samples for lometrexol analysis were collected for 24 h
`after drug administration at 6-hour intervals. and 20 ml aliquots
`
`Downloaded from clincanoerres.aacrjouma|s.org on February 1, 2011
`Copyright © 1995 American Association for Cancer Research
`
`Lilly Ex. 2048
`Sandoz v. Lilly IPR2016-00318
`
`

`
`M82 Lometrexol with Folic Acid Supplementation
`
`f.»“T
`
`/
`‘R
`
`0
`
`Time (min)
`
`or
`
`dS
`
`.5 O
`
`9 -A
`
`Lornetrexolconcentration(pgyrnl)3E;
`
`
`
`
`Lornetruttolconcentration(pglml)
`
`'P
`
`o
`
`1000
`
`3000
`2000
`'l1me (min)
`
`Fig. I Representative plasma disposition curves for total lometrexol in
`patients receiving folic acid supplementation: A. I2 mym‘ (A).
`l6
`mym’ (A): B. 30 mglm’ (O). 45 mglm’ (O) lometrexol. The lines are
`those generated by cornpamnental analysis.
`
`time points (224 h). plasma elimination could be described
`more accurately by a triexponential equation. The most appro-
`priate equations (biexponential or triexponential) fitted to each
`data set all had coefficients of detennination (:3) of >092. i.e..
`0.99 median (range. 0.92-0.99).
`Pharmacokinetic parameters are shown in Table 2 with
`parameters for patients receiving more than two courses of
`lomeuexol being represented as the mean and SD of all courses
`studied in that patient. Model-dependent analysis of lometrexol
`pharmacoltinetics resulted in a I'/:(! and MB of 19 t 7 min and
`256 : 94 min. respectively. and a my (where measurable) of
`H70 1' 435 min.
`
`Calculation of both lometrexol AUC and Cl-rm by either a
`model-independent or model-dependent analysis was not found
`to result in any significant difference (P = 0.l4 and P = 0.33.
`respectively. paired t test), and there were strong linear relation-
`ships for both parameters for both analyses (r > 0.98).
`Model-independent analysis
`indicated that
`lometrexol
`AUC was linearly related to dose (r = 0.88. P < 0.00l; Fig. 2).
`i.e.. plasma clearance of lometrexol was not dose dependent.
`However. two patients had a lometrexol AUC and Clm, which
`differed markedly from that observed in other patients treated at
`the same dose level. One patient (patient 12) receiving 30
`
`mglmz lometrexol had a consistently greater AUC than for
`others receiving the same dose. i.e.. 2.08-2.28 mg/ml - min
`compared with L38 and LSO mg/ml - min (Table 2 and Fig. 2).
`Reduced lometrexol clearance in this patient may have been due
`to an underlying early left ventricular failure. which could have
`decreased cardiac output and thereby reduced tissue perfusion.
`combined with a relatively low pretreatment GFR of 75 ml/min.
`which may have influenced the renal excretion of lometrexol. In
`contrast. at a dose of 45 mg/m" lometrexol. one patient (patient
`I9) had a much lower AUC than was measured in others (l.27
`mg/ml - min compared with values of 1.60-2.20 mg/ml - min).
`with a conespondingly higher Clm..- (35.8 ml/minim’ compared
`to values of 23.1-27.7 ml/minim’). This may be attributable to
`an unusually large lometrexol 'v'd,, in this patient of l5.3 liters!
`m‘'. which in turn could have been caused by the presence of a
`bilateral pleural effusion. combined with a high pretreatment
`GFR of I65 ml/min which would have promoted extensive renal
`excretion of lometrexol.
`
`No consistent change in the model-independent AUC was
`evident following more than one course of lometrexol. with the
`possible exception of three patients receiving 45 mg/mi lome-
`trexol every 4 weeks (patients 13. I6. and 17). in whom small
`increases in AUC after the second and third courses of lome-
`
`In patients receiving more than three
`trexol were observed.
`courses of lometrexol. no consistent change in clearance could
`be found. i.e.. the intrapatient CV of plasma clearance was l4%
`(median; range, 3—2l%; n = 5).
`The lometrexol Vd“ varied between 4.7 and lS.8 I/mi. and
`the unbound lometrexol Vd“. calculated from measurements of
`lometrexol plasma protein binding (see below). varied between
`l8.l and 67.1 iimz. The larger Vd». of two patients receiving 45
`mg/ml lometrexol may be attributed to the fact that both patients
`had pleural effusions. Even if data from these two patients are
`excluded. rank correlations are observed between dose level
`
`(mg/m2) and both Vdss (rs = 0.84. P < 0.00l)and unbound Vd,_
`(A = 0.66. P < 0.01).
`Plasma Protein Binding of Lometrexol. The percent-
`age of unbound lometrexol in plasma was 22 1 3%. indicating
`plasma protein binding of 78 t 3%. Protein binding detetrnined
`at two concentrations of lometrexol (3.5—4.7 and 7.7—2l.l ttgl
`ml) revealed that binding was not concentration dependent (P =
`0.! I. paired r test). An inverse linear relationship was apparent
`between the lometrexol binding and serum albumin concentra-
`tion (r = 0.88: P < 0.00l: Fig. 3). The relationship between
`unbound lometrexol and total serum protein was not found to be
`significant (r = 0.38; P > 0.05). when one patient with partic-
`ularly low total serum protein (55 g/liter) was removed from the
`analysis.
`Urinary Excretion of Lometrexol. The 0-24-h cumu-
`lative urinary excretion data (Fig. 4) for 19 patients (2! courses)
`indicated that the major elimination route for lometrexol was
`renal excretion. with 85 : l6% of the administered dose being
`excreted within 24 h of drug administration and 56 1 l7%
`within the first 6 h. Urinary excretion studied in two patients
`who received two consecutive courses of lometrexol did not
`
`reveal any consistent change in excretion following the second
`course of treatment. Two patients had 0-24-h lometrexol uri-
`nary recoveries of > l00% of the administered dose. which was
`likely to be due to inaccuracies in the measurement of urine
`
`Downloaded from clincanoerres.aacrjouma|s.org on February 1, 2011
`Copyright © 1995 American Association for Cancer Research
`
`Lilly Ex. 2048
`Sandoz v. Lilly IPR2016-00318
`
`

`
`Clinical Cancer Research I483
`
`Table 2 Pharmacultinetic parameters for lometrexol administered with folic acid supplementation
`
`PatienIIcour_se
`|ll—3
`ZI2
`3ll.2
`4/2-4
`5/I.2
`6/!
`8/2
`9ll.3.4
`l2ll—-1
`I3/I-3
`I411
`I5]!
`I6I|.2
`l7Il.2
`I8/I
`I9/I
`20/1
`Mean 1 SD
`(n = I7)
`
`Dose level
`
`(mg/m’)
`I2
`I2
`I2
`I6
`I6
`I6
`30
`30
`
`45
`45
`45
`45
`45
`
`45
`
`'11:“
`
`(min) _
`I2 : 2
`I9
`l8.24
`ll 1 2
`I2. [3
`I8
`I9
`I2 1' 3
`24: I9
`27: 6
`37
`9
`I4.24
`28.19
`I4
`22
`
`M25
`(min)
`224 I 31
`243
`276. 289
`222 I 38
`226. 235
`205
`282
`248 t 23
`295 t 246
`525 I 223
`392
`I23
`I60. 5 I 2
`265. 209
`I47
`I75
`I95
`
`'11:’?
`(min)
`
`I394 1 573
`
`795
`
`791
`l8ll
`I062
`
`AUC
`
`_ (mg/ml - min)
`0.57 I O.I2
`0.73
`0.57. 0.56
`0.77 I 0.09
`0.82. 0.83
`0.62
`l.52
`L38 I 0.I6
`2.|7 : 0.l I
`2.05 I 0.42
`2.20
`l.74
`l.60.
`I.77.
`2.13
`l.27
`1.79
`
`.
`.
`
`clearance
`(ml/min/m=_)__
`21.6 : 4.1
`16.8
`20.8. 21.4
`20.8 : 2.1
`19.1. 19.6
`26.0
`20.2
`22.0 1 3.0
`13.9 1 0.5
`22.5 I 4.7
`23.1
`27.2
`27.7. 22.5
`25.4. 20.8
`22.0
`35.8
`25.4
`
`tmllminl
`ma)
`|6.2
`
`l2.l
`
`14.5
`18.5
`
`22.5
`
`22.0
`19.7
`24.3
`27.7
`23. I
`I 5.6
`39.3
`29.5
`
`Vd
`,.
`tlitcrslmz)
`6.I I 0.5
`4.7
`6.7.6.8
`6.0 : 0.4
`5.5. 5.8
`6.1
`7.1
`7.2 : 0.2
`9.8 : 1.0
`10.8 : 3.2
`12.4
`12.0
`8.7. 10.8
`7.4. 8.8
`9.2
`15.8
`10.9
`
`256:96
`
`H70 1 435
`
`22.6 t 4.7
`
`2I.2 1' 7.5
`
`
`
`%Unboundlornetrexol
`
`2
`
`-A
`
`3 5 550:
`
`><
`
`40
`
`so
`
`20
`
`30
`
`40
`
`50
`
`Dose level (mg/m’)
`Fig. 2 Relationship between lomeuexol dose and AUC calculated
`fmm nnncnmpartmental analysis. Data are given in Table 2. and the line
`is that generated by linear regression analysis.
`
`Serum albumin (gll)
`Fig. 3 Relationship between serum albumin concentration and the
`percentage of lometrexol which was unbound to total plasma protein.
`The line is that given by linear regression analysis.
`
`volume. The 0-24-h urinary excretion of lometrexol was found
`to correlate linearly with the dose administered ( r = 0.69: P <
`0.01) even when these patients were omitted from the analysis.
`and a significant rank correlation was also evident (r, = 0.72;
`P < 0.001).
`Pharmncokinetics of Lometrexol Administered without
`
`Folic Acid Supplementation. Lometrexol phannacokinetics
`were studied in seven patients who received 45 or 60 mg/m’
`lometrexol without folic acid supplementation. A triexponential
`compartmental model was found to best describe plasma elim-
`ination of lometrexol in four of seven patients (Table 3) and a
`biexponential curve in the remainder. For most data sets a good
`curve fit was observed. the coefficient of determination (re)
`being greater than 0.89 in every case (median. 0.99; range.
`
`0.89-0.99). The mat and rv.-B determined from these analyses
`were I7 1' 8 min and I69 1 51 min. respectively. with a Iv.--y
`(where measurable) of 2593 t 1671 min. As was found with
`patients receiving folic acid. AUC and CIT." calculated using
`model-dependent analysis were not significantly different from
`those calculated using model-independent analysis (P = 0.30
`and P = 1.00). The lometrexol Vds, in these patients was highly
`variable. 116.. 5.7—28.0 Iiterslmz and 8.5—3I.3 liters/ml at 45 and
`60 mg/m3 lometrexol. respectively.
`in patients treated
`Plasma protein binding of lometrexnl
`without folate supplementation was virtually identical to that in
`patients receiving folic acid. with a value for the unbound
`lometrexol fraction of 22 1 4% and 78 1 4% for the protein
`bound fraction.
`
`Downloaded from clincanoerres.aacrjouma|s.org on February 1, 2011
`Copyright © 1995 American Association for Cancer Research
`
`Lilly Ex. 2048
`Sandoz v. Lilly IPR2016-00318
`
`

`
`
`
`\\\&\\\\\\\\\\\¢\\\\\\\\\\\ L
`
`
`
`$
`
`I484 Lometrexol with Folic Acid Supplementation
`
`5%
`\\\\\\\\\\
`
`;\\\\\\\\\\\\
`
`-Ak\\
`
`s
`
`|\\\\\\\\\\\\
`
`
`k\\\\\\\\\\\n\\
`
`\\\\\\\\\\uk\\\\\\\\\\\
`
`so
`
`
`
`\\\\\\\\\\VIR\\\\\\\\\\\§R\\\\\\\\\\\L\\
`
`
`12
`
`ii88
`
`CO
`
`03
`
`8 8
`
`
`
`
`
`Urinaryexcretion(96atdoseadministered)
`
`
`
`
`\\\\\\\\\\\k\\\\\\\\\\\\\\\\‘\\\\\\\\\\\\\\\\\\\\a\\\\\\\\\\\\\\‘k\\
`
`
`
`\\\\\\\\\\\\\\\\\\\\k\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
`
`
`
`The AUC in individual patients studied after multiple
`courses of lometrexol was found to be minimally cumulative in
`three patients and noncumuiative in five patients. Minimal ac-
`cumulation of lometrexol on repeated treatment was also de-
`scribed in the earlier Phase I study of Young et al. (5). Similarly.
`in patients receiving repeated Tomudex treatment. no accumu-
`lation is evident (l8). Lometrexol AUC was found to increase
`linearly with dose. an observation which is in agreement with
`data for other classical antifolates such as CB37l7 (N'°-propar-
`gyl-5,8-dideazafolic acid; Ref. 24). Tomudex (I8). and edatrex-
`ate (17). but not for methotrexate for which it has been sug-
`gested that there is a nonlinear relationship between dose and
`AUC following bolus administration (25).
`That there was a tendency for the lometrexol V6“ (and
`unbound Vd“) to be larger in patients receiving higher doses of
`the drug may be indicative of concentration-dependent protein
`binding. Thus. protein binding of lometrexol may be a saturable
`phenomenon, with greater drug concentrations resulting in a
`larger unbound fraction. Unbound lometrexol would be subject
`to rapid cellular uptake and hence an apparently larger Vd“.
`This possibility is supported by the observation that the fraction
`of lometrexol unbound is inversely related to serum albumin
`concentration (Fig. 4). A relationship between Vd“ and dose has
`also been observed in pharmacokinetic studies with Tomudex in
`rats (26).
`The Vd“ range (4.7-15.8 liters/m’) from patients treated
`with l2—45 mg/m2 lometrexol was. on average. smaller than
`that measured in patients who did not receive folate supplemen-
`tation. but were treated with 45 and 60 mg/m’ lometrexol
`(5.7-3l.8 liters/m2). and that reported for patients treated in a
`previous study (i4—32 iitersimz) with doses of l5-60 mg/"ml
`(27). This observation may again reflect the possible concentra-
`tion-dependent plasma protein binding of lometrexol and the
`effect that higher lometrexol concentrations would therefore
`have on drug disposition.
`Despite these findings. the in virro protein-binding data did
`indicate that concentration-dependent protein binding of
`not
`lometrexol was statistically significant (P > 0.05) between the
`concentration ranges of 3.5 and 4.7 and 7.7 and 21.0 uglml.
`although a comparison of binding by a paired Student's I test
`resulted in a P value of 0.ll. which would indicate a trend
`toward reduced lometrexol binding at higher concentrations.
`The magnitude of lometrexol protein binding (78%) was
`comparable to that of methotrexate (28. 29). but not with the
`thymidylate synthasc inhibitors CB37l7 and Tomudex which
`bind more extensively (97% and >90%; Refs. 24 and 26). Since
`lometrexol has a high affinity for membrane-bound FBP (30.
`31), it should be noted that a component of the protein binding
`measured in this study may have involved binding to soluble
`FBP. Soluble FBP present in plasma is thought to function as a
`folate transport protein (32) and has a M, 35.000—|00.000 (33).
`which is above the threshold (30.000) used to detennine protein
`binding.
`Any condition influencing lometrexol protein bin