[CANCER RESEARCH 50, 3493-3502, June 15, 1990]
`
`Relationship between Dose Rate of [6RS]Leucovorin Administration, Plasma
`Concentrations of Reduced Folates, and Pools of 5,10-Methylenetetrahydro-
`folates and Tetrahydrofolates in Human Colon Adenocarcinoma Xenografts
`
`Janet A. Houghton,2 Larry G. Williams, Siebold S. N. de Graaf, Pamela J. Cheshire, John H. Rodman,
`Dan C. Maneval, Irving W. Wainer, Philippe Jadaud, and Peter J. Houghton
`
`Laboratories for Developmental Therapeutics, Department of Biochemical and Clinical Pharmacology [J. A. H., L. G. W., P. J. C., P. J. H.], and Department of
`Pharmacokinetics [J. H. R., D. C. M., I. W. ;V., P. J.], St. Jade Children’s Research Hospital, Memphis, Tennessee 38101, and Department of Pediatrics [S. S. N. d. G.],
`University Hospital, P. O. Box 30.001, 9700 RB Groningen, The Netherlands
`
`ABSTRACT
`
`by FUra, and FUra cytotoxicity in colon tumors under the in situ
`conditions of tumor growth.
`
`[6RS]Leucovorin (5-formyitetrahydrofolate; 5-CHO-H4PteGlu) ad-
`ministered in different regimens in combination with 5-fluorouracil
`(FUra) has increased the response rates to FUra in patients with colon
`adenocarcinoma. Using preclinical models of human colon adenocarci-
`nomas as xenografts in immune-deprived mice, the effect of the rate of
`administration of racemic [6RS]leucovorin on the concentration-time
`profile of reduced folates in plasma, size of intratumor pools of 5,10-
`methylenetetrahydrofolates (CH2-H,PteGIu.) and tetrahydrofolates
`(H4PteGlu.), and the distribution of their polyglutamate species have
`been examined.
`Bolus injection i.v., or 4-h or 24-h infusion of [6RS]leucovorin (500
`mg/m2) yielded similar concentration profiles of the biologically active
`[6S] and inactive [6R] isomers of 5-CHO-H4-PteGlu and 5-methyl-
`tetrahydrofolate (5-CH3-H4PteGIu) in mouse plasma to those previously
`reported in humans, but with more rapid elimination half-lives (t~ = 11
`to 16 min, 23 to 41 min, and 30 to 35 min, respectively). Thus, reduced
`folates remained elevated in plasma during the period of [6RS]leucovorin
`administration. In HxELC, and HxGC3 tumors, pools of CH2-H4PteGIu.
`and l-L~PteGlu, were increased from 350% to 700% of control, but only
`during [6RS]leucovorin infusion. Intracellular levels subsequently de-
`clined rapidly, similar to the loss of reduced folates from plasma. Increas-
`ing the rate of [6RS]leucovorin delivery by decreasing the time for
`administration from a 24-h to a 4-h infusion did not further increase the
`intratumor pools of CH2-H4PteGIu. and H4PteGlu., suggesting saturation
`in the cellular metabolism of [6RS]leucovorin.
`In HxGC3 tumors, CH2-H4PteGiu4-s were elevated more rapidly than
`in line HxEI~2, which accumulated predominantly a shorter chain length
`species following i.v. bolus injection. During the 4-h infusion schedule,
`di- and triglutamate species in particular accumulated in both tumors
`with no elevation in CH~-l-14PteGlus until the infusion was discontinued,
`when this species increased as the shorter chain length forms were
`declining. However, during the 24-h infusion of [6RS]leucovorin, CHz-
`H4PteGlu~-s were elevated in tumors. Since these species have been
`reported to increase the binding affinity of [6-3H]5-fluorodeoxyuridine
`monophesphate ([6-3H]FdUMP) to thymidylate synthase, and intratumor
`pools of CH~-H4PteGlu. and H4PteGlu. were elevated during the 24-h
`infusion of [6RS]leucovorin, this was considered to be the preferred
`schedule for administration. When FUra (6.25 to 25 mg/kg) was admin-
`istered 3 h into a 24-h infusion of [6RS]leucovorin (500 mg/m~) in tumor-
`bearing mice, potentiation of thymidylate synthase inhibition in compar-
`ison with FUra administered alone was observed. These studies raise
`important questions regarding the effect of (a) dose of [6RS]-
`leucovorin, (b) frequency of administration, (c) utility of 5-CHs-
`H4PteGla, and (d) [6R]5-CHO-H4PteGlu on influencing intratumor pools
`of CTI2-1~PteGlu. and H4PteGlu., the inhibition of thymidylate synthase
`
`Received 8/4/89; revised 3/8/90.
`The costs of publication of this article were defrayed in part by the payment
`of page charges. This article must therefore be hereby marked advertisement in
`accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
`~ Supported by National Cancer Institute Awards CA32613, CA23099, and
`Cancer Center Support (CORE) Grant CA21765; Center of Excellence Award,
`University of Tennessee; and American Lebanese Syrian Associated Charities.
`2 To whom requests for reprints should be addressed.
`
`INTRODUCTION
`
`5-Fluorouracil is one of the most effective single agents used
`in the treatment of adenocarcinoma of the colon. However,
`responses to the agent administered singly either in patients (1,
`2) or to xenografted tumors in mice (3) have been transient.
`Previous studies from these laboratories suggested that, in
`human colon adenocarcinoma xenografts, concentrations of
`CH2-H4PteGlun3 were suboptimal to allow maximal formation
`or stability of the covalent ternary complex formed between
`thymidylate synthase, CH2-H4PteGIun, and the metabolite of
`FUra, FdUMP (4-6). Data also suggested that it would be
`advantageous to administer a reduced folate with FUra in vivo
`to increase the pools of CH2-H4PteGIun in colon tumors (4), in
`particular, concentrations of the longer polyglutamate chain
`length forms. These species have been shown to increase the
`affinity of binding of FdUMP to the enzyme at concentrations
`lower than required for binding in the presence of the mono-
`glutamyl form (6).
`Leucovorin, which is a mixture of diastereoisomers and a
`stable form of reduced folate, has subsequently been used in
`clinical trials in combination with FUra in the treatment of
`patients with colon adenocarcinoma. This strategy was based,
`in part, upon the xenograft studies and also upon in vitro studies
`using cultured cells that demonstrated a 1.5- to 4.6-fold poten-
`tiation of FUra- or FdUrd-induced cytotoxicity by [6RS]-
`leucovorin (7-11). In several independently conducted Phase
`III randomized clinical trials, FUra in combination with
`[6RS]leucovorin has shown significant increases in response
`rates over FUra administered alone (3- to 5-fold) in the treat-
`ment of colon adenocarcinomas (12-16). In the combination
`arms, a significant increase in time to disease progression (12,
`13, 15), increase in patient survival (13, 15), and increased
`therapeutic index (15) over FUra administered alone have been
`reported. Of interest is that [6RS]leucovorin has been admin-
`istered by i.v. bolus injection daily for 5 days (13, 15), by a
`short duration of infusion (2 h) weekly (14, 16), or by continu-
`ous infusion over 5 to 6 days (12) at dose levels of 20 to 500
`mg/m2, each of which has resulted in significant increases in
`response rates to FUra.
`Plasma pharmacokinetics of the individual isomers of leu-
`covorin and the major metabolite of the biologically active
`[6S]leucovorin, 5-CH3-H4PteGIu, has also been reported for
`several clinical regimens (16-19). However, no data, either
`
`3 The abbreviations used are: CH2-HoUteGlu., 5,10-methylenetetrahydrofol-
`ates; 5-CHO-H~PteGIu, 5-formyltetrahydrofolate ([6RS]leucovorin); FUra, 5-
`fluorouracil; l-UPteGlu., tetrahydrofolates; 5-CH3-H4PteGlu, 5-methyltetrahy-
`drofolate; FdUMP, 5-fluorodeoxyuridine monophosphate; FdUrd, 5-fluorodeox-
`yuridine; HPLC, high-pressure liquid chromatography; SDS, sodium dodecyl
`sulfate.
`
`3493
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1110-0001
`
`

`
`LEUCOVORIN METABOLISM IN COLON TUMORS
`
`clinical or preclinical, are available that determine how the
`plasma concentrations of these reduced folate forms and their
`maintenance relate to the elevation and maintenance of intra-
`tumor concentrations of CH2-H4PteGIu~ or how [6RS]leuco-
`vorin influences the distribution of the polyglutamate species
`or the inhibition of thymidylate synthase by FUra. Conse-
`quently, we have examined these relationships in preclinical
`models of human colon adenocarcinomas maintained as xeno-
`grafts in immune-deprived mice. A dose of [6RS]leucovofin
`(500 mg/m2) was chosen to provide plasma concentrations
`similar to those used clinically (16-18) and was administered
`to mice i.v. by bolus injection, short infusion (4 h), and pro-
`longed infusion (24 h). The effects of dose rate of [6RS]-
`leucovorin administration on the plasma concentration-time
`profiles of reduced folates in mice and, subsequently, on intra-
`cellular pools of CH2-H4PteGlun and H~PteGlun in two human
`colon xenograft lines were evaluated. In addition, the 24-h
`infusion schedule that was considered to be optimal was sub-
`sequently selected for examination of the effect of [6RS]-
`leucovorin (500 mg/m~) on FUra-induced thymidylate synthase
`inhibition.
`
`MATERIALS AND METHODS
`
`Chemicals. [5-3H]dUMP (specific activity, 20 to 22 Ci/mmol) and
`[6-3H]FdUMP (specific activity, 18 to 20 Ci/mmol) were obtained from
`Moravek Biochemicals, Brea, CA. Glycine was obtained from Eastman
`Kodak Co., Rochester, NY, and Tris (ultrapure) from Boehringer
`Mannheim Biochemicals, Indianapolis, IN. Polyacrylamide, sodium
`dodecyl sulfate, and all other reagents for gel electrophoresis were
`purchased from Bio-Rad Laboratories, Richmond, CA. Pteroylpolyglu-
`tamates (for preparing CH2-H4PteGlun; 6) were obtained either from
`Dr. Charles Baugh, University of South Alabama College of Medicine,
`Mobile, AL, or from American Radiochemicals, Inc., St. Louis, MO.
`Lactobacillus casei thymidylate synthase (specific activity, 1.8 to 8.2
`~mol/h/mg; specific activity, 72.4 to 211 ~mol/h/ml; 1 unit converts 1
`~mol of dUMP/h to dTMP) was purchased from Biopure, Boston,
`MA. Charcoal (activated, neutralized) was supplied by Sigma Chemical
`Co., St. Louis, MO. NCS tissue solubilizer and ACS scintillant were
`obtained from Amersham Corp., Arlington Heights, IL; PPO and
`POPOP were from Research Products International Corp., Elk Grove
`Village, IL; and En3hance was from New England Nuclear, Boston,
`MA. Kodak XoOmat AR film was purchased from Eastman Kodak Co.
`HPLC-grade methanol was obtained from American Scientific (Mu-
`skegon, MI). gBondapak phenyl HPLC columns were purchased from
`Waters Chromatography Division, Milford, MA, and columns of bo-
`vine serum albumin bound to 10-~m spherical silica (Resolvosil) were
`from Rainin Instruments, Woburn, MA. [6R]I0-CHO-H4PteGlu was
`kindly provided by Dr. John McGuire, Roswell Park Memorial Insti-
`tute. All other reagents were supplied by Sigma or were of reagent
`grade.
`Tumor Lines. Human colon adenocarcinomas HxELC2 and HxGC3
`were maintained as xenografts by passage in the s.c. space of female
`CBA/CaJ mice (Jackson Laboratories, Bar Harbor, ME) immune-
`deprived by thymectomy, followed by potentially lethal whole-body
`irradiation and reconstitution with syngeneic bone marrow as described
`previously (6). Line HxELC2 has shown some sensitivity to 5-fluoro-
`pyrimidines after treatment of tumor-bearing mice, whereas HxGC3
`tumors were intrinsically resistant to these agents in vivo (3). A 5-fold
`difference in thymidylate synthase activity has been determined in the
`2 tumor lines, where the activity was 0.186 and 0.991 nmol/g of tissue/
`min, respectively, determined from the release of~H from [5-~H]dUMP
`(20).
`Plasma Pharmacokinetics of [6S]Leucovorin, [6RILeucovorin, and
`[6SIS-CH~-H~PteGlu in Mice. One hundred seven non-tumor-bearing,
`immune-deprived mice received [6RS]ieucovorin at a dose level of 500
`mg/m2 by i.v. bolus injection or by 4-h or 24-h infusion through a
`cannula introduced into the tail vein. For infusions, [6RS]leucovorin
`3494
`
`was delivered from a Harvard Model 975 infusion pump at a rate of
`0.14 ml/h. The relationship between mg/kg and mg/m2 was calculated
`for individual mice, as previously described (21). Blood samples were
`obtained at regular intervals for 360 min after bolus injection (n -- 31),
`at 2 time points during and 4 time points following the 4-h infusion (n
`-- 35), and at 7 time points during and 60 min following the 24-h
`infusion (n -- 41). For each mouse, a single blood sample of 0.3 to 0.8
`ml was obtained by cardiac puncture under metafane anesthesia. Sam-
`ples were collected in heparinized syringes and placed on ice, and
`sodium ascorbate was added immediately to a final concentration of 1
`mg/ml. Samples were centrifuged at 2000 × g for 10 rain at 2"C.
`Plasma was removed and stored at -70°C prior to analysis.
`A coupled HPLC achiral-chiral assay system was used to separate
`and quantitate the individual enantiomers of [6RS]leucovorin and 5-
`CH3-H4PteGlu using methotrexate as an internal standard. The enan-
`tiomeric mixture was quantitated on the achiral column and subse-
`quently separated on a chiral system as previously described (22).
`Detection limits were 100 ng/ml (=0.2 ~M) for 5-CHO-H~PteGIu and
`5-CH3-H,PteGIu. Retention times on the chiral column were 28 and
`40 min for the [6S] and [6R] enantiomers of the parent compound and
`58 min for the [6S]leucovorin metabolite, 5-CH3-H4PteGIu (22).
`Pharmacokinetic parameters for a one- and two-compartment model
`were estimated using maximum likelihood with each observation
`weighted according to the inverse of the variance for the model estimate
`assuming a coefficient of variation of 10% (23). Model selection was
`based on comparative examination of residuals and the weighted sum
`of squares. Data for each of the enantiomers of [6RS]leucovorin for the
`bolus, 4-h, and 24-h infusions were analyzed separately. For the metab-
`olite 5-CH3-H4PteGlu, pharmacokinetic parameters were determined
`using a parent-metabolite model. The independently estimated [6S]-
`leucovorin parameters were fixed as a driving function with the as-
`sumptions that 80% of the parent compound was converted to metab-
`olite by a first order process. Varying the assumption of parent to
`metabolite conversion from 50 to 100% did not influence the estimate
`of metabolite plasma half-life.
`Infusions of 16RS]Leucovorin in Tumor-bearing Mice. Mice bearing
`HxELC2 or HxGC3 tumors were infused i.v. with [6RS]leucovorin by
`bolus administration or by infusion for 4 h or 24 h. At various times
`during or after infusion, mice (2 per point) were killed. Tumors (2 per
`mouse) were rapidly excised, pooled (unless otherwise stated), and
`placed immediately in and subsequently stored in liquid nitrogen.
`Pooled tumors were ground to a fine powder under liquid nitrogen, and
`the extracted powders were used to examine the modulation of pools
`of CH2-H~PteGIun and H4PteGlun. Alternatively, pooled tumors were
`allowed to thaw to 2"C on ice prior to examination of the effect of
`[6RS]leucovorin on FUra-induced thymidylate synthase inhibition.
`Determination of Pool Size of CH2-H4PteGlu. and 1-14PteGlu.. Due to
`the instability of CH2-H4PteGlu~ to heat treatment in the absence of
`excess HCHO (5, 24), endogenous concentrations of the combined
`pools of CH2-H4PteGIu. and H4PteGlu. were determined in the pres-
`ence of excess HCHO that converted 1-14PteGlun to CH2-H4PteGIun.
`The assay was based on the catalytic release of ~H from [5-3H]dUMP
`over 3 min, where the rate of reaction was determined to be independent
`of the glutamyi form of the cofactor (5). Reactions were linear over 3
`min under the conditions used. L. casei thymidylate synthase and tumor
`extracts (containing 1% ~-mercaptoethanol and 10 mM sodium ascor-
`bate) as the source of reduced folates were used in reaction mixtures as
`described previously (5). Endogenous dUMP was removed by treatment
`of extracts with 5’-nucleotidase prior to assay. Batches of thymidylate
`synthase were examined for formylase activity (25) using [6R]I0-CHO-
`H4PteGlu and, where necessary, were further purified on columns of
`Scphadex G-100 and CM-Sephadex (26) that removed the contaminat-
`ing enzyme.
`Distribution of Polyglutamates of CH:-H4PteGIu. Determination of
`the predominance of CH2-H4PteGIu polyglutamates in HxGC3 and
`HxELC2 tumors following [6RS]leucovorin administration was based
`upon the technique described by Priest and Doig (27), where the
`distribution of CH2-H4PteGiu. had been previously characterized in
`untreated tumor xenografts (5). This method has been currently used
`as a qualitative technique only, with quantitation of pool size expansion
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1110-0002
`
`

`
`LEUCOVORIN METABOLISM IN COLON TUMORS
`
`by the 3H release assay described above. Ternary complexes formed
`among [6-3H]FdUMP (I 25 riM), excess L. casei thymidylate synthase,
`and CH2-H4PteGIun from tumor extracts were electrophoresed on 9%
`polyacrylamide nondenaturing gels (28 cm), and fluorograms were
`prepared. Equivalent volumes of reaction mixtures containing 10%
`glycerol (120 or 160 ~l) were applied to each gel. A modified gel-
`processing procedure yielding improvement of sample reproducibility
`was used for some of the studies. The new procedure involved fixation
`for 1 h at room temperature in a mixture of glacial acetic acid (10%,
`v/v) and methanol (30%, v/v) in water, followed by treatment with
`En3hance for 1 h with gentle agitation. Following impregnation, gels
`were agitated in an excess of cold water for 30 min (one change) and
`were subsequently dried. Data were analyzed by scanning densitometry
`of fluorograms to determine the intensity of hands. The relationship
`between peak height and dpm was linear (r2 = 0.944) over the range
`examined. Each experiment was controlled internally by electrophoresis
`of untreated and [6RS]leucovorin-treated samples on the same gel.
`SDS-Gel Analysis of [6-3H]FdUMP-Thymidylate Synthase-CH~-
`tL~PteGlu, Complexes. Tumor extracts (200 ~l) prepared for the deter-
`mination of size of pools of CH2-H,PteGlun and H4PteGlun were
`incubated at 37"C for 30 min with L. casei thymidylate synthase (1.8
`units), [6-~H]FdUMP (125 riM), and 25 mM Tris-HCl (pH 7.4) contain-
`ing O-mercaptoethanol (1%) and sodium ascorbate (10 mM). Mixtures
`were subsequently heated at 100"C for 3 min with 50 ~l of SDS sample
`buffer (5×) to denature covalent ternary complexes and were stored at
`-70"C until analyzed. Ninety or 180 ,I of each sample were applied to
`12% polyacrylamide-SDS gels (1.5 mm x 14 cm x 16 cm) with a 4%
`polyacrylamide-SDS stacking gel (4 cm). Ternary complexes were
`electrophoresed at 25 mA/gel according to the method of Laemmli
`(28). The method was used to qualitatively determine the appearance
`and disappearance of covalent ternary complexes during and following
`modulation by [6RS]ieucovorin.
`Inhibition of Thymidylate Synthase. The effect of a 24-h i.v. infusion
`of [6RS]leucovorin (500 mg/m2) on the inhibition of thymidylate syn-
`thase by FUra in HxELC2 and HxGC~ tumors was examined. A
`suboptimal dose of FUra was selected that would cause incomplete
`inhibition of thymidylate synthase at the nadir, followed by some
`recovery by the end of the infusion, such that potentiation by [6RS]-
`leucovorin may be more readily determined. The dose of FUra used in
`mice bearing HxELC2 tumors was 12.5 mg/kg and for HxGC~ was
`6.25 or 25 mg/kg. FUra was administered i.v. by bolus injections 3 h
`into a 24-h infusion of [6RS]leucovorin or saline, by means of a 3-way
`valve, and was flushed through the cannula with 0.1 ml of saline (0.9%).
`At this time, plasma levels of reduced folates were approaching a steady-
`state concentration, and intratumor pools of CH2-H4PteGlun and
`H,PteGlu, were elevated. At 4 h and 21 h after FUra administration,
`tumors were excised (4 tumors pooled; 2 mice per time point), cytosols
`were prepared, and endogenous nucleotides were removed by charcoal
`adsorption procedures. The rate of release of ~H from [5-3H]-
`dUMP was subsequently determined in each sample and compared to
`the rate of release of ~H in control tumors, as described (20, 29). Data
`were analyzed statistically using a one-way analysis of variance.
`
`Plasma Concentration-Time Profiles for Reduced Folates
`
`[6R] isomer (555 ~tM and 649/zM, respectively) when sampling
`commenced at 5 min. The [6S] isomer was rapidly eliminated,
`with a monoexponential t~ of 11.4 min (Table 1), and was
`undetectable at 2 h after injection. For [6R]leucovorin, elimi-
`nation from plasma was more prolonged and biexponential (t~
`= 18.2 min; t~ = 41.2 min). For the major metabolite of [6S]-
`leucovorin, 5-CH3-H4PteGlu, appearance in plasma was rapid,
`with the maximal concentration (34 ~M) achieved at approxi-
`mately 30 min following [6RS]leucovorin bolus administration.
`Its estimated elimination t~ from plasma (30.3 min) was inter-
`mediate between the values for the two enantiomers of [6RS]-
`leucovorin. At 100 min after injection, the ratios of the concen-
`trations of [6R]leucovorin to [6S]leucovorin and of 5-CH3-
`H4PteGlu to [6S]leucovorin were estimated to be 11 and 8,
`respectively.
`Four-h Infusion. During a 4-h infusion of [6RS]leucovorin,
`lower levels of all reduced folates were observed in plasma (Fig.
`1; Table 1). At 3 h during the infusion, concentrations of [6S]-
`leucovorin, [6R]leucovorin, and 5-CH3-tLPteGIu were deter-
`mined to be 55, 78, and 7.7 ~M, respectively. Upon cessation
`of infusion, each was eliminated from the plasma, with half-
`lives of 8.7, 23.2, and 34.9 min, respectively. The pharmacoki-
`netic parameters estimated from the bolus administration ex-
`periments with [6RS]leucovorin did not reliably predict [6S]-
`leucovorin and 5-CH3-H4PteGIu concentrations following the
`4-h [6RS]leucovorin infusion (Fig. IA), suggesting nonlinearity
`in disposition at differing dose rates. Thus, the t~ for elimina-
`tion from plasma was estimated using only the postinfusion
`data (Fig. 1,4, - .... ). The data for [6R]leucovorin during and
`following the 4-h infusion were, however, adequately described
`by a linear pharmacokinetic model.
`Twenty-four-h Infusion. When the duration of infusion of 500
`mg/m2 of [6RS]leucovorin was extended to 24 h, lower concen-
`trations of [6S] (5.1 ~M) and [6R] (16 ~M) leucovorin and 5-
`CH3-I-L~PteGIu (3 /tM) were determined. Concentrations of
`[6S]ieucovorin approached maximum values within 1 to 3 h
`from the initiation of the infusion, but did show a tendency to
`accumulate during the infusion. Reduced folates remained ele-
`vated in plasma during the infusion and were again rapidly
`eliminated after discontinuing the infusion. A one-compart-
`ment linear pharmacokinetic model was used to describe the
`data for each reduced folate during and following the 24-h
`infusion of [6RS]leucovorin. The t~ values for elimination of
`[6S]leucovorin and [6R]leucovorin following the infusion were
`15.6 and 35.3 min, respectively, consistent with data derived
`following the alternate two [6RS]leucovorin administration reg-
`imens. Although insufficient data were available to accurately
`estimate a t~ for elimination of 5-CH3-H4PteGIu from plasma
`following the 24-h [6RS]leucovorin infusion, the pattern of
`metabolite accumulation during the infusion was consistent
`with an elimination t~ of 35 min determined following a 4-h
`infusion.
`For all [6RS]leucovorin infusion regimens, where plasma
`concentrations of reduced folates had been determined in tu-
`mor-bearing mice, these were similar to those reported for the
`concentration-time profile study determined in non-tumor-
`bearing mice (data not shown).
`
`Concentrations of reduced folates achieved in plasma follow-
`ing bolus administration or 4-h or 24-h infusion of [6RS]-
`leucovorin (500 mg/m2) were examined. The data and com-
`puter-simulated plasma concentration-time profiles for [6S]-
`leucovorin and 5-CH3-H~PteGIu are shown in Fig. 1,4 and for
`[6R]leucovorin in Fig. IB for each of the 3 different adminis-
`tration regimens. The maximal plasma concentrations achieved
`and the half-time (t,~) for elimination from plasma for each
`reduced folate and schedule are summarized in Table 1.
`In HxELC2 and HxGC3 tumors, modulation of the pools of
`Bolus Administration. After i.v. bolus injection of [6RS]-
`CH2-H4PteGIu~ and H4PteGlu~, as determined by the catalytic
`leucovorin, plasma concentrations of the biologically active
`release of 3H from 3
`[5- H]dUMP, followed the maintenance and
`[6S] isomer of leucovorin were lower than for the inactive
`disappearance of the potentially biologically active reduced
`3495
`
`Determination of CH2-I-I~PteGIu. and H4PteGlu. Pools in Neo-
`plastic Tissues
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1110-0003
`
`

`
`Fig. I. Plasma concentration-time profiles of
`[6S]leucovorin (O) and 5-CH3-H,-PteGlu (I:]) (,4) _
`and [6R]leucovorin (&) (B) in mice following i.v. ~
`bolus administration (top), 4-h infusion (center), ~
`or 24-h infusion (bottom) of [6RS]leucovorin (500
`mg/ma). Plasma samples were analyzed and data
`evaluated according to procedures described in
`"Materials and Methods." Data pertaining to
`[6$]leucovorin and 5-CH3-H,PteGlu for the 4-h
`infusion schedule were fit to the linear pharmaco-
`kinetic model (--) or, alternatively, data ob-
`tained following the infusion were analyzed by
`linear regression ( .... ).
`
`!
`~-
`
`LEUCOVORIN METABOLISM IN COLON TUMORS
`
`100o
`
`lO
`
`0.1
`
`lOO
`
`0,1
`
`A lOOO
`
`B
`
`ao ~0 eo ~0
`T~me
`
`too
`
`~20
`
`o.1
`~40 0
`
`80
`
`100 150 200 250 300
`
`IO0
`
`0O
`
`IO0
`
`~5O 200
`Time (nm)
`
`25O
`
`0O0 ~
`
`0
`
`5O
`
`100
`
`~5O 20O
`~me (rain)
`
`±SO 5O0 ~0
`
`100
`
`0.1
`
`o
`
`200
`
`400
`
`600
`
`800
`
`1000 1200 1400 1600
`
`200 400 600
`
`800 1000 1200 1400 1800 1800
`
`Time (m~n)
`
`Time (mkl)
`
`Table 1 Maximal plasma concentration~ of reduced folates achieved in mice and t~ for elimination following different schedules of [6RS]leucovorin administration
`
`Maximal plasma concentration (~M)a t~ (min)
`
`Schedule
`
`[6S]Leucovorin
`
`[6R]Leucovorin
`
`5-CH~-H~PteGlu
`
`[6S]Leucovorin
`
`[6R]Leucovo6n
`
`5--CH3-H,~PteGlu
`
`i.v. bolus
`4-h infusion
`24-h infusion
`
`555 + 110~ (4)c
`55 ± 10 (8)
`5.1 ± 3.1 (32)(cid:128)
`
`649 + 113 (4)
`78 + 10 (8)
`16 ± 7 (31)e
`
`34 + 2 (2)
`7.7 - 1.6 (8)
`3.0 ± 1.0 (11)e
`
`11.4
`8.7
`15.6
`
`41.2d
`23.2
`35.3
`
`30.3
`34.9
`NEf
`
`"Determined from derived data.
`~ Mean ± SD.
`(cid:128) Numbers in parentheses, number of data points.
`’~ Terminal half-life estimated by fitting data to a two-compartment model. Initial t~ = 18.2 min.
`¯ Average value from all data derived between 3 and 24 h representing an average steady-state concentration.
`fNE, not evaluable.
`
`ing mice (Fig. 2), intratumor pools of CHa-H,PteGIu. and
`H,PteGlu~ were elevated to 253% and 344% of control at 12
`and 24 h, respectively, in HxGC3 tumors, and to 677% and
`702% of control at these times in line HxELCa. However, these
`pools rapidly declined following the end of infusion, as plasma
`levels of reduced folates also decreased.
`
`folates ([6S]leucovorin and 5-CH3-H,PteGIu) from plasma.
`Following an i.v. bolus injection of [6RS]leucovorin, pools of
`CH~-H4PteGIu, and H4PteGlu~ were elevated in both tumor
`lines to 242% (HxGC3) or 409% (HxELC2) of control at 1 h
`after injection (Fig. 2), the earliest time point examined. After
`this time, intratumor reduced folate pools declined rapidly,
`approaching control levels by 6 h after [6RS]leucovorin admin-
`istration.
`The influence of [6RS]leucovorin (500 mg/m2) on the distri-
`At the end of a 4-h infusion of [6RS]leucovorin in mice
`bution of polyglutamate species of CHa-H4PteGIu and the
`bearing HxELC~ tumors, the combined pools of CH2-
`combined pools of CH2-H,PteGIu and H~PteGlu was subse-
`H4PteGlu~ and H4PteGlu, were elevated to 660% of control in
`quently examined in both tumor lines for the 3 different rates
`tumors and were observed to return to basal levels at 4 h after
`of administration. Within a tumor line, qualitatively similar
`the infusion had ceased (Fig. 2). In line HxGC3, intratumor
`effects were observed for the two pools for each dose rate of
`reduced folate pools increased to 253% of control 2 h into the
`[6RS]leucovorin administration. Consequently, data for the
`4-h infusion and were maintained at 221% of control at the end
`CH2-H4PteGIu~ pool alone have been presented.
`of infusion. However, as had been observed in HxELC: tumors
`HxGC~ Tumors. In HxGC3 tumors, there was a predominance
`following infusion, intratumor concentrations of CH2-
`of penta- and hexaglutamate species (Figs. 3,4 and 4,4). Follow-
`H~PteGlu~ and H4PteGlu~ rapidly declined, following the elim-
`ing bolus injection of [6RS]leucovorin (500 mg/m2) to tumor-
`ination of reduced folates from plasma (Fig. 2).
`bearing mice, there was a rapid appearance of CH2-I-I~PteGIu4
`During the 24-h infusion of [6RS]leucovorin in tumor-bear-
`3496
`
`Distribution of CH:-H4PteGlu Polyglutamates in Tumors
`
`Sandoz Inc. IPR2016-00318
`Sandoz v. Eli Lilly, Exhibit 1110-0004
`
`

`
`LEUCOVORIN METABOLISM IN COLON TUMORS
`
`A. Bolus
`
`HxGC
`
`HxGC3
`
`24 hr
`
`40O
`
`300
`
`200
`
`100
`
`0
`
`6
`
`12 18 24 30
`
`800
`
`HxELC 2
`
`400
`
`200
`
`2 hr 4 hr 8 hr
`
`B. 4 hr infusion
`
`0
`
`6
`
`12 18 24 30
`
`Time (hours)
`
`Fig. 2. The influence of [6RS]leucovorin (500 mg/m2) administered by i.v.
`bolus injection, 4 h or 24 h infusion, on the size of pools CHrH,PteGlu, and
`H~PteGlu, in HxGC3 and HxELC2 tumors. The assay was based on the release
`of 3H from [5-3H]dUMP as described in =Materials and Methods." Points, mean
`of 12 to 16 determinations on 4 individual tumors at each time point; bars, SE.
`
`4 hr
`0
`~ intusion ~
`
`5 hr 8 hr
`
`C. 24 hr infusion
`
`0 24hr 26hr 30hr
`~ infusion~
`
`Fig. 3. The distribution of polyglutamate species of CH:-H~PteGlu in HxGC~
`tumors following i.v. bolus administration (A), during and following a 4-h infusion
`(B), or 24 h infusion (C) of [6RS]leucovorin (500 mg/m~) in tumor-bearing mice.
`[6-~H]FdUMP-thymidylate synthase-CH~-H~PteGlu, complexes, formed using
`tumor extracts, were electrophoresed on 9% nondenaturing gels, which were
`subsequently fixed in 5% trichloroacetic acid (A) or 10% glacial acetic acid:30%
`methanol in water (B and C) prior to treatment with En3Hance. Gels were further
`treated and dried, and fluorograms were prepared as described in ~Materials and
`Methods."
`
`and increased formation of CH2-H4PteGIu5, within 2 h. Tetra-,
`penta-, and hexaglutamate species continued to predominate 8
`h after initial administration of [6RS]leucovorin; a trace of
`CH2-H4PteGlu2 was detected at 4 h.
`At the end of a 4-h infusion of [6RS]leucovorin in mice
`bearing HxGC3 tumors, the distribution of CH2-H4PteGIu,
`differed from that observed after i.v. bolus injection (Figs. 3B
`and 4B). It was evident in HxGC3 tumors that, during the
`infusion, there was marked accumulation of CH2-H4PteGIu2
`and CH2-H~PteGlu3, some increase in CH2-H4PteGIu~, and a
`concomitant decrease in CH2-H4PteGIus and rapid disappear-
`ance of CH2-H4PteGIut. By 1 h after the end of the infusion,
`CH2-H4PteGIu3_5 predominated, and after a further 3 h, di- and
`triglutamate species were declining. CH2-H4PteGIu6 could not
`be detected during and after [6RS]leucovorin infusion, although
`a small quantity of the hexaglutamate was detected within the
`shorter chain length form declined by 8 h after injection, CH2-
`combined pool of CH2-H4PteGIu, and H~PteGlu, (data not
`H4PteGlu4 and CH2-H4PteGlu5 became elevated.
`shown). Thus, during the 4-h infusion of [6RS]leucovorin,
`At the end of the 4-h infusion of [6RS]leucovorin (Fig. 5B),
`accumulation of shorter polyglutamate chain length forms of
`the predominant change observed in HxELC2 tumors was sim-
`CH2-H4PteGIu was detected in HxGC3 tumors that was not
`ilar to that observed in HxGC3, in that accumulation of CH:-
`observed following bolus administration of [6RS]leucovorin.
`H4PteGlu2 was evident, with some elevation in the tri- and
`By the end of the 24-h infusion of [6RS]leucovorin, CH2-
`tetraglutamates; CH2-H4PteGlu5 was observed to decrease.
`H~PteGlu~_5 were the predominant species in line HxGC3, while
`Upon cessation of infusion, however, CH2-H~PteGlu2 declined,
`CHa-H4PteGIu2 was not substantially elevated. CH2-H4PteGIu6
`with a concomitant increase in CH2-H4PteGIu3, CH2-H4PteGIu4,
`was again not detectable in HxGC3 following treatment of
`and CH2-H4PteGIus. By 4 h postinfusion, the tetra- and pen-
`tumor-bearing mice with [6RS]leucovorin. For up to 6 h after
`taglutamates were the predominant species.
`the end of the infusion, CH2-H4PteGIu3_5 predominated (Figs.
`With the 24-h infusion of [6RS]leucovorin (Fig. 5C), CH2-
`3C and 4C).
`H4PteGlu3 and CH2-H4PteGIu4 were most markedly elevated
`HxELC: Tumors. After i.v. adminis