APPLIED
`PHARMACOKINETICS
`
`Principles of Therapeutic Drug Monitoring
`
`
`Edited by
`
`William E. Evans, Pharm.D.
`Director, Clinical Pharmacokinetics Laboratory
`and Clinical Division of Pharmacy
`-
`St. Jude (_3hildren’s Research Hospital
`and
`
`Associate Professor of Clinical Pharmacy
`Department of Pharmacy Practice
`University of Tennessee
`Center for the Health Sciences
`Memphis
`
`Jerome J. Schentag, Pharm.D.
`Assistant Director, Clinical Pharmacokinetics Laboratory
`Millard Fillmore Hospital
`and
`
`Assistant. Professor of Pharmaceutics and Pharmacy
`State University of New York. at Buffalo
`Buffalo
`
`William J. Jusko, Ph.D.
`Director, Clinical Pharmacokinetics Laboratory
`Millard Fillmore Hospital
`and
`Professor of Pharmaceutics
`State University of New York at Buffalo
`Buffalo
`
`Applied Therapeutics, Inc.
`Spokane, WA
`
`Elie
`
`-
`
`'
`
`
`
`MEDAC Exhibit 2013
`
`ANTARES v. MEDAC
`
`IPR2014-01091
`
`Page 00001
`
`MEDAC Exhibit 2013
`ANTARES v. MEDAC
`IPR2014-01091
`Page 00001
`
`

`
`Other publications by Applied Therapeutics, Inc:
`
`DRUG INTERACTIONS NEWSLETTER: A Clinical Perspective and
`Analysis of Current Developments, Edited by Philip D. Hansten.
`ISSN 0271-8707
`
`Basic Clinical Pharmacokinetics by Michael E. Winter.
`ISBN 0-915486-04-0
`
`Applied Therapeutics: The Clinical Use of Drugs, Edited by Brian
`S. Katcher, L1oyd'Yee Young, Mary Anne Koda—Kimb1c.
`ISBN 0-915486—05—9
`
`Copyright © 1980 by Applied Therapeutics, Inc.
`Printed in the United States of America
`
`All rights reserved. No part of this book may be reproduced, stored in a
`retrieval system, or transmitted, in any form or by any means, electronic,
`mechanical, photocopying, recording, or otherwise without prior written
`permission from the publisher.
`'
`
`Applied Therapeutics, Inc.
`P.0. Box 1903
`
`Spokane, WA 99210
`
`Library of Congress Catalog Card Number 80-53408
`ISBN 0-915486—03—2
`
`Second Printing—April 1981
`Third Printing—July 1983
`
`Page 00002
`
`Page 00002
`
`

`
`16
`
`Methotrexate
`
`William E. Evans, Pharm.D.
`
`INTRODUCTIONFBACKGROUND
`
`Methotrexate (MTX, amethopterin, 4-amino-N”-methyl pteroylg1u-
`tamic acid) is an analogue of aminopterin, the folic acid antagonist
`introduced in 1948 by Farber for the treatment of acute leukemia.
`MTX is a weak acid with a pK£_ in the range of 4.8 to 5.5 (1) and
`differs from aminopterin by being methylated at the N” position.
`MTX exerts its cytotoxic effects by competitively inhibiting dihydro-
`folate reductase, the intracellular enzyme responsible for converting
`folic acid to reduced folate cofactors.
`
`I1
`
`Absorption
`
`Despite the number of years that MTX has been used and the
`number of oral doses that have been given, relatively little is known
`about its absorption from the gastrointestinal tract. There have been
`several publihed studies describing the oral absorption of MTX; how-
`ever, they either included a small number of patients, used a non»
`specific assay or both. Nevertheless, these studies have provided the
`basis for current recommendations regarding oral dosing. The gas-
`trointestinal absorption of MTX is considered to be dose dependent,
`with low dosages of MTX (<30 mgfm2) reported to be well absorbed (2,
`3) while the extent of absorption is reduced to 50 to 70% with doses
`in excess of 80 mglmz (3). However, absorption characteristics at these
`higher dosages (2 80 mg/m”) are based on observations in one patient
`administered 80 mga’m"(3) and one patient given 10 m-gfkg (-300 mgfm“)
`(2). These data are also difiicult to interpret because of the non-specific
`assay methods used.
`The data regarding oral absorption at the lower dosages are based
`on a somewhat larger patient population, but do not necessarily in-
`dicate that the drug is completely absorbed. Re-evaluation of pub-
`lished data from two patients given 3 mgfmz (2), 28 patients given 15
`
`518
`9!;
`
`Page 00003
`
`Page 00003
`
`

`
`....._._
`
`______
`
`Methotrexate I
`
`519
`
`mgfmz (4,5) and seven patients given 30 mglmz (3) indicates that the
`percentage of absorption was 83%, 57 to 69% and 4.7% respectively.
`Most of these studies also used non-specific assays, making precise
`interpretations difficult. In two patients given 3 mgfm”, $ 9% of the
`administered dose was recovered in a three day stool collection while
`39% of the dose was recovered in the stool following 10 mg/kg (~300
`mg.-’m2) given to one patient. In a study (3) where chromatographic
`analysis was used, approximately 47% of the administered dose (30
`mglm‘) was recovered in feces as MTX and 35% as a metabolite, while
`6% was recovered as metabolite after the same dosage given intra-
`venously. This suggests that metabolism occurs during the absorption
`process from the gastrointestinal tract, which is supported by the
`studies of Valerino (6), who reported that MTX is metabolized by
`intestinal bacteria to 4-amino-4—deoxy-N“’—methylpteroic acid (DAMPA).
`The intestinal metabolite (DAMPA) has about 1f200th the affinity of
`MTX for dihydrofolate reductase (DHFR), the target enzyme. This
`metabolite is also produced during enterohepatic circulation of sys-
`temically administered MTX (6). In children given oral MTX in des-
`ages of 20, 30 and 40 mgfmg every 6 hours for four doses, we found
`significantly higher MTX concentrations following 30 or 40 mg/m‘’
`versus 20 mgfmg, but no significant difference between 30 and 40
`mghn” (7). Animal studies (8) support the dose dependent nature of
`MTX absorption, with the process being best describedusing Michae-
`lis~Menten kinetics with Km and Vmax values of 1.5 X 10-5 M and
`4.8 X 10"? Mfmin, respectively. There is also substantial variability
`in the time of peak concentrations following oral doses (7,9), and the
`rate of absorption has been reported (5) to decline during a six week
`course of therapy.
`The administration of a four drug regimen of oral non-absorbable
`antibiotics has been reported (5) to reduce the oral absorption of MTX
`(15 mgfmz) from 69% to 44%, while similar findings have also been
`described following pretreatment with oral neomycin alone (10). This
`is apparently the result of malabsorption secondary to the oral anti-
`biotics. I-Iowever, Shen and Azarnoff (10) report that pretreatment
`with systemic kanamycin leads to a substantial increase in the plasma
`level and recovery of intact MTX after oral administration, presum»
`ably due to reduction in MTX metabolism by gut bacteria.
`
`Distribution
`
`Following intravenous administration, MTX distributes within an
`initial volume approximating 18% (0.18 Lfkg) of body weight (11) and
`exhibits a steady-state volume of approximately 75 to 80% of body
`
`
`
`Page 00004
`
`Page 00004
`
`

`
`520 1' Methotrexate
`
`weight (2,12). Methotrexate is approximately 50% bound to plasma
`proteins (1,2,3), primarily albumin, at serum concentrations ranging
`from 10“‘ M to 10"3 M, (0.5 to 50 p.g/ml). Other protein-bound organic
`acids such as salicylates, sulfonamides and para aminohippurate can
`displace MTX from protein binding sites (1,13). The clinical signifi-
`cance of these drug-drug interactions is difficult to assess, since many
`of these organic acids may also competitively inhibit renal tubular
`secretion of MTX (1).
`
`In animals (14) and man (15) the highest tissuefplasma equilibrium
`distribution ratios are reported for the kidney and liver, followed by
`the gastrointestinal tract and muscle. The gastrointestinal tract is
`apparently an important site of distribution and metabolism of both
`orally and intravenously administered methotrexate. Zaharko (14)
`and Bischoff (15) in studies of MTX disposition in lower animals and
`man reported the persistence of higher MTX concentrations in gut
`lumen of the small intestine when compared to liver, kidney, muscle
`or plasma. Their model predictions in mouse and man indicated that
`higher plasma levels in man are due to less rapid clearances by the
`kidney and bile and a longer residence time in the human small
`intestine. Methotrexate in the gastrointestinal tract may either be
`reabsorbed by a saturable process, excreted in the feces, or taken up
`and metabolized by bacteria in the large intestine. The differences
`between man and smaller animals in the persistence of MTX in the
`gastrointestinal tract, and the rate and extent of metabolism by gut
`bacteria are due in part to differences in transit time in the small and
`large intestine (14). Decreased GI transit rate secondary to complete
`or partial gastrointestinal obstruction has been described as a poten-
`tial mechanism for delayed total body clearance of MTX in man
`(16.17).
`Following the administration of high doses of MTX, distribution of
`MTX into pleural fluid or ascites may also have a substantial influence
`on MTX total body clearance (18,19). The presence of a pleural effusion
`resulted in a significant increase in the terminal phase half-life in a
`patient extensively evaluated with and without a pleural effusion
`(19). The presence of a pleural effusion resulted in significant de-
`creases in the disposition rate constant and the K21 intercompartment
`distribution rate constant of a two-compartment first-order kinetic
`model. Beginning six hours after
`the high~dose methotrexate
`(HDMTX) infusion, MTX concentrations in pleural fluid were always
`greater than the simultaneous serum concentrations. These data sup-
`port the hypothesis that patients with ascites or pleural effusions are
`at increased risk for developing toxicity following HDMTX because of
`
`l
`
`Page 00005
`
`Page 00005
`
`

`
`Methotrexate X
`
`521
`
`delayed MTX clearance. The maximum concentration of MTX in I
`pleural effusions and ascitic fluid is only about 10% of the maximum
`serum concentration, but declines more slowly with an eventual
`pleural fluid/‘serum equilibrium ratio of approximately 10. This makes
`the influence of pleural effusions or ascites most significant when high
`doses (>50 mgfkg) of MTX are given. When such high doses are given
`in patients with "third spaces,” MTX accumulated in these extravas-
`cular compartments acts as a source of “sustained release” and delays
`total body clearance of MTX. Such patients are at greater risk of
`having potentially cytotoxic MTX concentrations beyond the usual
`duration of leucovorin rescue. Although the effect of pleural effusion
`on the decline in serum concentrations may not be evident until 24 to
`30 hours after the dose, serum concentrations at this time are still
`approximately 100—fold greater than the minimum concentration re-
`quired for inhibition of DNA synthesis (~10"8 M).
`An important consideration in the distribution of MTX is the process
`of membrane transport, since the effects of MTX are dependent upon
`an intracellular concentration sufficient to inhibit dihydrofolate re-
`ductase activity. The mechanism of membrane transport of MTX has
`been extensively studied and reviewed by Goldman and associates
`(20). Simplistically, intracellular transport can occur by two processesi
`simple transmembrane diffusion and a carrier—mediated active trans-
`port process. At relatively low extracellular MTX concentrations (i.e.
`10“‘ M), the active transport process predominates. This active trans-
`port process follows Michaelis-Menten kinetics, with the rate of influx
`proceeding at half the maximum transport velocity when extracellular
`concentrations are approximately 5 X 10”‘ M (19). This Michaelis
`constant is similar to values reported for naturally occurring reduced
`folates. Following the intravenous administration of high doses of
`MTX (2 100 mgfkg), serum concentrations in the ranges of 10“ to 10 ‘3
`M are achieved. At these concentrations, the active transport process
`is saturated and passive diffusion becomes a major pathway by which
`effective intracellular concentrations can be achieved. This may be of
`particular importance in the treatment of malignant diseases which
`have an acquired or de novo resistance to MTX due to a reduced active
`transport process. Additionally, since MTX and reduced folates (i.e.
`leucovorin) share the same active transport process, high extracellular
`MTX concentrations can reduce or inhibit the intracellular transport
`of leucovorin (21). Thus, leucovorin “rescue” following HDMTX is a
`competitive rescue, despite its non-competitive biochemical mecha-
`nism of circumventing the inhibition of DHFR with reduced folates.
`When MTX serum concentrations are ~10"? M, MTX effects can be
`
`
`
`Page 00006
`
`Page 00006
`
`

`
`
`
`522 I Methotrexate
`
`rescued with equimolar serum concentrations of leucovorin, while
`10‘3 M concentrations of leucovorin may be required when MTX
`concentrations are IO‘5 M (21).
`
`Using L1210 leukemia cells in-vitro, Zager et a1 (22) reported that
`transmembrane influx and efflux of MTX may also be affected by other
`drugs, including vincristine, corticosteroids, asparaginase and ceph~
`alothin. Using Ehrlich ascites tumor cells, Fyfe and Goldman (23)
`further characterized this interaction, demonstrating that 1 X 10-”
`M vincristine (VCR) slowed the efliux of MTX. It was initially proposed
`that the inhibition of MTX efflux by vincristine could be exploited
`clinically to produce higher intracellular concentrations of MTX. How-
`ever, more recent studies using human leukemia cells (24) indicate
`that VCR concentrations of 1 X 10' '5 M and 1 X 10"’ M enhance the
`
`intracellular accumulation of MTX by 54% and 33%, respectively.
`More important, however, was their observation that if VCR-incu-
`bated cells were washed free of VCR and resuspended in VCR-free
`media, the inhibition of MTX efflux is lost. Thus, VCR must be present
`(in suificient concentrations) for enhancement of MTX intracellular
`
`accumulation to be observed. Owellen and coworkers first reported on
`the pharmacokinetics of VCR in man in 1976 (25,26). Their studies
`demonstrated that the peak concentrations of VCR achieved after
`bolus intravenous administration of dosages tolerated clinically were
`approximately 1 X 10-7 M. Moreover, these concentrations were very
`short-lived with a t1/2,1,,“ of 3.37 i 0.72 minutes and a t1/zbm of 155
`:18 minutes. More recently, Warren and coworkers (27) failed to
`demonstrate a significant enhancement of intracellular MTX accu-
`mulation in human lymphoblastoid cells when the clinically achiev-
`able VCR concentration of 1 X 10'? M was used. This was noted for
`extracellular MTX concentrations achieved after either conventional
`
`or high-dose MTX administration.
`The blood-cerebrospinal fluid (CSF) barrier is relatively impermeable
`to MTX (2,28,29) and CSF concentrations following intravenous
`administration of MTX are dose dependent. Animal studies indicate
`that the brain:serum ratio for MTX after low dosages (10 mglkg) is
`small (i.e. 0.11) (30,31). In humans, CSF MTX concentrations have
`been reported to range from 10”’ M after a 24-hour infusion of 500
`mg.-‘ma to greater than 10"'5 M following 7500 mg/m2 given as an
`intravenous bolus (32,33). The distribution of MTX from plasma into
`CSF is apparently slow, with two recent reports demonstrating that
`peak CSF concentrations are reached only toward the end of 24-hour
`infusions of 500 or 1000 mgfm” (34,35). Studies conducted in a small
`number of patients indicate that lumbar CSF and brain extracellular
`
`Page 00007
`
`Page 00007
`
`

`
`Methotrexate I 523
`
`fluid concentrations are similar following intravenous high-dose MTX
`administration (36). This is not the case when MTX is given intrathe-
`cally into the lumbar CSF, since MTX distribution into ventricular
`CSF is variable and unpredictable following intrathecal administra-
`tion.
`Ionizing radiation has been shown to alter the permeability of the
`blood-brain barrier to MTX, although these changes may vary from
`one region of the brain to another and from one species to another.
`Griffin and associates (30) reported that no MTX was detectable in
`brains of unirradiated mice following 100 mgfkg intraperitoneal MTX,
`while detectable MTX levels were present after 2000 rads cranial
`irradiation, but not after 500, 1000 or 1500 rads. These findings are
`consistent with the suggestions of previous clinical investigators that
`MTX-related leukoencephalopathy is most often a result of combined
`therapy with cranial irradiation and systemic MTX, based on the
`postulate that radiation alters the integrity of the blood-brain barrier,
`allowing MTX to diffuse more easily into the white matter (2,37,38)'.
`
`Metabolism
`
`-
`
`As previously stated, MTX is metabolized by intestinal bacteria to
`4-amino-4-deoxy- 1°-methylpteroic acid (DAMPA) and other minor
`metabolites (6). The DAMPA metabolite has about 1!200th the affinity
`of MTX for the target enzyme, DHFR (6).
`Another potentially important metabolite of MTX is its oxidation
`(aldelhyde oxidase) product, 7-hydroxy methotrexate (7-OH MTX)
`(39). This metabolite is reported to constitute 1 to 11% of the admin-
`istered dose recovered in a cumulative 24-hour urine collection follow-
`ing high-dose intravenous MTX. Because 7-OH MTX is about two
`orders of magnitude less effective as an inhibitor of DHFR (40) and
`because it represents such a small percentage of the total dose found
`in the urine, it was initially considered to be a relatively unimportant
`metabolite. However, more recent studies (41) report serum concen-
`trations of '?'-OH MTX exceeding concurrent serum concentrations of
`the parent drug, following high-dose (200 mgfkg) intravenous admin-
`istration of MTX. This finding (which has recently been confirmed)
`coupled with the fact that 7-OH MTX is three- to five-fold less water
`soluble than MTX (39) makes this a potentially important metabolite,
`since the urine concentrations may exceed the solubility of the parent
`compound -and metabolite (42). The aqueous solubility of MTX is pH
`dependent, and can be reduced from 10 Inglml to 1 mgiml by reducing
`the pH from 6.9 to 5.7 (42). MTX urine concentrations exceeding 5
`
`1
`
`:
`
`
`
`Page 00008
`
`Page 00008
`
`

`
`
`
`524 1‘ Methotrexate
`
`mgfrnl (10"“ M} have been reported (42), and urine pH’s less than 6.0
`are frequently observed following high doses of MTX (without urinary
`alkalinization). Thus, intratubular precipitation leading to obstruc-
`tive nephropathy has been suggested as one mechanism of MTX renal
`toxicity (43). For these reasons, alkalinization of the urine prior to
`and following high-dose MTX infusions has become standard practice.
`Animal studies have shown that MTX may be converted to mono-
`and diglutamate forms in Vivo (44). Whitehead and coworkers (45)
`have reported that MTX monoglutamate was formed immediately
`after administration and then disappeared with a half-life of 6.5 days.
`Poly-3'-glutamates have been reported in red blood cells of a patient
`being treated with MTX (44). The fact that poly-y-glutamates can be
`converted back to MTX by hydrolase enzymes suggests that they may
`serve as an intracellular reservoir for MTX. However, their presence
`in human tissues and the pharmacological importance of the gluta-
`mate metabolites remain to be determined.
`
`Excretion
`
`Renal excretion is the major route of methotrexate elimination,
`constituting greater than 80% of total body elimination (42,46—49).
`Estimates of MTX clearance in adults with normal renal function
`
`yield values of-about 110 mlfminfm” (1). MTX clearance in these pa-
`tients exceeded glomerular filtration rate {inulin clearance) by 6 to
`49%, suggesting active tubular secretion. Following intravenous in-
`fusion of high doses of MTX, greater than 40% of the administered
`dose has been recovered unchanged in urine within 6 hours and 90%
`within 24 hours (2,46,50,51). However, these values are somewhat
`greater than those described by Isacoff et al (52) who reported only
`60% cumulative urinary recovery following 64 high-dose infusions.
`The reason for these discrepancies may involve the specificity of the
`assays used, although this does not explain all of the observed differ-
`ences. Cumulative 24-hour urinary recovery of lower intravenous
`doses of MTX (0.1 to 10 mgfkg) is reported to be 58 to 92% (median
`78%). As previously stated, urinary recovery following oral adminis-
`tration is lower and dose dependent, reflecting gut metabolism and the
`incompleteness of absorption.
`Net renal clearance of MTX has been reported to decrease from
`78 :5 mlfmin when serum concentrations range from 10' '7 to 10' '5 M,
`to 25 to 50 mlfmin at serum levels ranging from 10”“ M to 10"3 M
`(53). The low net renal clearance at lower serum levels suggests
`extensive tubular reabsorption of MTX, while the decreasing renal
`clearance at higher serum concentrations indicates that tubular se-
`
`Page 00009
`
`Page 00009
`
`

`
`Methotrexate ; 525
`
`o’
`
`-
`
`cretion of MTX may be saturated at concentrations attained clinically.
`Renal tubular secretion of MTX can be competitively inhibited by
`other organic acids such as salicylates and sulfonamides (1), and can
`be totally blocked by probenecid (54). However, the reduction in total
`body clearance due to saturation of the active tubular secretory pro.
`cess or competitive inhibition by organic acids is seldom clinically
`significant.
`The time course of MTX disappearance from plasma following high-
`dose intravenous infusions is essentially biexponential (-16,17,42,52,55).
`Mean half-lives for the initial phase have been reported to range from
`1.5 to 3.5 hours (42,49,50), while the terminal phase half-life is around
`8 to 15 hours in patients with normal total body clearance. Many of
`the early studies reporting half-lives were not designed as pharma-
`cokinetic studies, and the reported initial-phase half-lives represent
`the rate of decline in serum concentrations (during the first 24 hours)
`and not t‘/2,..,,1.a i0.693!alpha). However, Isacoff and coworkers (52) fit
`serum concentrations following 172 high-dose infusions to a biexpo~
`nential equation and calculated parameters of a two—compartment
`model, yielding a t];/galpha of 1.8 i 0.5 hours. These values are similar to
`those reported by Stoller et al (42) who also used a biexponential model.
`No significant diiferences in kinetic parameters were observed at dos-
`ages ranging from 50 to 200 mgfkg (52). The early distribution half-
`life observed after IV bolus administration (11,49) is usually not ob-
`served following IV infusions, since distribution in the central com-
`partment essentially occurs during administration. The terminal phase
`half~life of 27 hours reported by Huffman et al (49) is longer than
`reported by other investigators and is most likely a result of the non-
`specific assay which measured total radioactivity of both MTX and
`metabolites (56). Importantly, the terminal phase half-life appears to
`correlate best with toxicity (12), as will be discussed later in detail.
`There are currently data (57) which suggest that the pharmacoki-
`netics of HDMTX may be significantly different in children. Wang and
`coworkers (57) recently reported significantly lower 6 and 24 hour
`serum MTX concentrations in children S 10 years of age compared to
`adults. Pharmacokinetic parameters derived from a very small num-
`ber of subjects (three children and six adults) indicated a shorter
`alpha
`11%
`and larger Vd in children. Urinary recovery of MTX in children
`was also greater during the infusion when compared to adults. How-
`ever, there was considerable overlap in serum "concentrations in both
`age groups at all time points and absolutely no difference at 48 and
`72 hours post-infusion. Moreover, the significant differences in con-
`centrations at 6 hours may not be clinically important. It seems rea-
`sonable for children with creatinine clearances greater than adults to
`
`
`
`Page 00010
`
`Page 00010
`
`

`
`
`
`526 r’ Methotrexate
`
`excrete MTX more rapidly; however, these potential age-related dif~
`ferences remain to be clearly defined.
`Because the large amount of MTX excreted in urine (after high
`doses) may exceed the 2 mM solubility at pH 5.5, vigorous hydration
`and urinary alkalinization have been recommended to prevent MTX
`precipitation and nephrotoxicity. The use of vigorous intravenous hy-
`dration (> 100 mlfmzfhour) does not appear to alter the plasma dispo-
`sition curve of MTX when compared to the same patients given main-
`tenance IV hydration (~40 mlfmzfhour).
`(58) The maintenance of
`urinary pH 7 (using oral sodium bicarbonate) for 12 hours before and
`48 hours after high—dose MTX significantly reduces the risk of renal
`toxicity (33) and is currently recommended for all patients (59,60).
`Active biliary secretion of MTX probably occurs (53,61), but is a
`relatively minor excretory pathway. The total amount recovered in
`the gastrointestinal tract following intravenous administration is less
`than 10% of the administered dose (49,53,62). Shen and coworkers
`(53) have reported bile-plasma concentration ratios ranging from 63
`to 145 in six cancer patients. The contribution enterohepatic circula-
`tion makes to the long terminal half-life of MTX remains undefined.
`However, this relatively minor excretory pathway (<10% fecal excre-
`tion) may become clinically important when impaired (i.e. GI obstruc-
`tion) (16,17).
`
`Intrathecal Methotrexate
`
`. Intrathecal (IT) administration of MTX is currently a standard
`component of preventive CNS therapy for childhood acute lymphocytic
`leukemia. Intrathecal MTX is also frequently used to treat active
`CNS involvement of leukemia, lymphoma and other responsive ma-
`lignancies. The disposition of IT MTX in the cerebrospinal fluid (CSF)
`is difficult to assess because of the need for repeated lumbar punctures
`to obtain serial CSF samples from individual patients. For this reason,
`very few studies report pharmacokinetic parameters derived from
`serial CSF concentrations. Assessment of MTX CSF disposition is
`further complicated by the uneven distribution of IT MTX between
`lumbar and ventricular CSF, and the intrinsic difficulties in obtaining
`ventricular CSF samples. Despite these limitations, some potentially
`useful information about the CSF disposition of IT MTX is available.
`Bleyer and coworkers (63) measured lumbar CSF MTX concentra-
`tions in 76 patients given IT MTX, with serial samples assessed in
`five of these patients. All patients were given the same dosage of 12
`mgfmg, in an injection volume of 12 mlfm” up to a maximum of 18 ml.
`None of these patients had evidence of active CNS disease. When the
`
`Page 00011
`
`Page 00011
`
`

`
`;_
`
`Methotrexate I
`
`527
`
`lumbar CSF MTX concentrations measured at various times (12 to 96
`hours) after an IT dose (for 76 patients) were collectively used to
`simulate CSF disposition, a biphasic disappearance curve was pro-
`duced. Half-lives of 4.5 hours and 14 hours were estimated during the
`intervals of 4 to 36 and 48 to 96 hours after the injection, respectively.
`These two half-life values were similar to values estimated from serial
`concentrations measured in one patient with chronic meningeal leu-
`kemia. Concurrent MTX concentrations in plasma (resulting from the
`IT dose) reached a peak of about 10'? M between 3 and 12 hours after
`the IT injection, and declined in parallel with the CSF concentrations.
`The terminal CSF half-life of MTX appears to be longer (64) in pa-
`tients with active meningeal leukemia, which is consistent with pre-
`vious observations that CSF MTX concentrations and the likelihood
`
`ofneurotoxicity are greater in patients with active meningeal leuke-
`mia andfor meningeal carcinomatosis (28,65). The disappearance of
`MTX from lumbar CSF is apparently dependent upon a number of
`physiologic processes including: [1] bulk flow removal via normal path-
`ways of CSF absorption, [2] bulk flow distribution within the subarach-
`noid and ventricular CSF, [3] diffusion throughout the ECF of the
`brain parenchyma and spinal cord, [4] diffusion from the ECF into the
`capillaries -of the brain and spinal cord and [5] absorption from ven-
`tricular fluid by the energy-dependent transport process of the choroid
`plexus (66). It has been hypothesized that delayed clearance of MTX
`from the CSF of patients with active meningeal disease is due to
`impairment of bulk flow removal of MTX. The rate of decline in MTX
`CSF concentrations can also be prolonged by either probenecid (67) or
`vincristine (68).
`
`The dose of intrathecal MTX is usually based on the patient’s body
`surface area (12 mgfmg). However, the CNS volume of children older
`than 3 years approaches that of adults, while body surface area does
`not plateau at adult levels until 16 to 20 years of age. Thus, CSF
`concentrations of MTX after IT administration of 12 mgfm” are gen-
`erally higher as age increases from 3 to 20 years (64). This may
`partially explain the increased risk of neurotoxicity in older patients
`given IT MTX (65). It has been proposed that all patients greater than
`3 years of age be given the same dose of MTX (12 mg) and not a
`dosage based on body surface area (64). The clinical use of CSF MTX
`concentrations to modify therapy will be discussed later in greater
`detail.
`'
`
`Methotrexate may also be given intraventricularly, and at least one
`study (28) has indicated that the distribution of methotrexate in CSF
`is more reliable when the drug is administered intraventricularly via
`an indwelling intraventricular subcutaneous reservoir compared to
`
`\
`
`
`
`Page 00012
`
`Page 00012
`
`

`
`528 1' Methotrexate
`
`intrathecal administration. However, the distribution of intrathecal
`MTX in this study was sufficient to achieve therapeutic levels in the
`ventricular CSF and may have been significantly better had the vol-
`ume of the intrathecal preparation been larger. Rieselbach et al (69)
`reported that the volume of the injected solution is an important factor
`in attaining widespread distribution following intrathecal adminis-
`tration. When the volume of injected solution is -10% of the estimated
`cerebrospinal fluid volume, adequate distribution is obtained only at
`the level of the basal cisterns; whereas when the volume is approxi-
`mately 25% of the estimated CSF, distribution is obtained throughout
`the cerebral subarachnoid space and ventricular system (28). Al-
`though subarachnoid and ventricular distribution of methotrexate
`may not be critical when cranial irradiation ‘is given concomitantly
`for CNS prophylaxis, good cerebrospinal fluid distribution is essential
`if methotrexate alone is employed for the treatment or prophylaxis of
`CNS leukemia.
`
`CONCENTRATION VERSUS RESPONSE AND TOXICITY
`
`Therapeutic Concentrations
`
`The cytotoxic effects of MTX are a result of its competitive inhibition
`of the intracellular enzyme dihydrofolate reductase (DHFR). The K,
`for this inhibition has not been precisely defined, although estimates
`of 10“" M have been made (70). Extensive studies by Goldman (71,
`72) and others have demonstrated that a free intracellular MTX con-
`centrations in excess of that required to saturate the tight binding
`sites on DHFR are necessary for maximal suppression of DNA syn-
`thesis. It appears that only a small fraction of uninhibited DHFR is
`sufficient to maintain reduced folate pools adequate to sustain DNA
`synthesis (71), thus necessitating intracellular concentrations of free
`MTX in excess of DHFR to maintain inhibition of the biochemical
`
`pathway. There is no feasible means by which intracellular MTX
`concentrations can be routinely measured in clinical specimens. While
`the relationship between extracellular and intracellular MTX concen-
`trations has been determined for several experimental tumors (7346)
`and for intestinal mucosa (75) it remains to be clearly established for
`most tissues. Since extracellular drug is in rapid exchange with in-
`tracellular free drug in sensitive cells, it seems reasonable that extra-
`cellular drug concentrations might relate to intracellular drug effects.
`Animal studies have indicated that the inhibition of DNA synthesis
`in tumor cells, bone marrow and intestinal epithelium requires the
`presence of a serum concentration of free MTX specific for each tissue
`
`Page 00013
`
`Page 00013
`
`

`
`Methotrexate I
`
`529
`
`(75). The inhibition of DNA synthesis in mouse bone marrow is vir-
`tually complete with plasma MTX concentrations above “-10-” M,
`whereas intestinal epithelium shows similar inhibition at MTX levels
`above 5 X 10"" M. Subsequent studies using an infusion device to
`maintain constant serum concentrations demonstrated the partial in-
`hibition of DNA synthesis at levels of 2 X 10”’ M and more complete
`inhibition of intestinal mucosa at this concentration (7 6,77). Similar
`findings have been reported in humans, where resumption of DNA
`synthesis did not occur until serum concentrations were 2 X 10"“ M
`or below (78). Pinedo and Chabner (79) have shown that at MTX
`_ concentrations >10‘“ M the cytotoxic effects are a function of both
`drug concentration and duration of exposure. Their data demonstrated
`that exposure to extracellular concentration of 5 X 10*‘ -M for 72
`hours produces the same effect as exposure to 10‘5 M for 12 hours. It
`therefore seems reasonable to assume that extracellular concentra-
`tions less than 10”‘ M are not likely to produce pharmacologic or
`toxicologic ef