Pharmacokinetics of Subcutaneous Methotrexate
`By Frank M. Balis, Joseph Mirro, Jr, Gregory H. Reaman, William E. Evans, Cynthia McCully,
`Karen M. Doherty, Robert F. Murphy, Susan Jeffries, and David G. Poplack
`
`The pharmacokinetics of subcutaneously adminis-
`tered methotrexate was studied as a parenteral al-
`ternative to oral administration. An initial feasibil-
`ity study was performed
`in Rhesus monkeys
`comparing the subcutaneous route to intravenous
`(IV) injection and oral administration. The sub-
`cutaneous dose was completely absorbed and a
`sustained-release effect was observed when com-
`pared with the IV dose. No local or systemic toxici-
`ties resulted from subcutaneous methotrexate in
`the animals. Twelve children with acute lympho-
`blastic leukemia on maintenance therapy protocols
`prescribing either 7.5 mg/m 2 biweekly or 40 mg/m2
`weekly were also monitored after both a subcuta-
`neous and an oral dose of methotrexate. Four chil-
`dren at the higher dosage level were also studied
`after an equal IV dose. The subcutaneous dose was
`
`NTERMITTENT
`low-dose methotrexate
`has become standard therapy for the mainte-
`nance of remission in children with acute lym-
`phoblastic leukemia (ALL) and is also used as
`an immunosuppressant in the treatment of a
`variety of other conditions, including psoriasis,
`rheumatoid arthritis, and asthma. At doses of 30
`mg/m2 and less, methotrexate is routinely ad-
`ministered by the oral route. However, pharma-
`cokinetic studies have demonstrated that plas-
`ma methotrexate concentrations following oral
`administration are highly variable as a result of
`interpatient differences in the rate and extent of
`absorption.'4 In one study, peak levels occurred
`
`Froni the Pediatric Branch, National Cancer Institute, Be-
`thesda, MD; St Jude Children's Research Hospital, Memphis;
`and Childrens Hospital National Medical Center, Washington,
`DC.
`Submitted June 17, 1988; accepted July 26, 1988.
`Supported in part by NIH Grants No. CA20180 and
`CA21765. DrMirro is a recipient of a Clinical Oncology Career
`Development Award from the American Cancer Society.
`Address reprint requests to Frank M. Balis, MD, Pediatric
`Branch, NCI, Bldg 10, Room 13N240, 9000 Rockville Pike,
`Bethesda, MD 20892.
`This is a US government work There are no restrictions on its
`use.
`0732-183X/88/0612-0006$0.00/0
`
`again completely absorbed in these children at
`both dose levels, whereas the oral dose, which pro-
`duced comparable plasma drug concentrations at
`the lower dosage level, resulted in a total drug ex-
`posure (area under the plasma concentration-time
`curve) that was one third that of the equal subcuta-
`neous dose at the higher dosage level. No local or
`systemic toxicity was attributed to the subcutaneous
`methotrexate. Subcutaneous administration of
`methotrexate is well tolerated and well absorbed
`and appears to overcome the problems associated
`with oral administration, including variable ab-
`sorption and saturation of the absorption mecha-
`nism with increasing doses.
`J Clin Oncol 6:1882-1886. This Is a US government
`work. There are no restrictions on Its use.
`
`from 0.5 to five hours after the dose, and the
`fraction of the dose absorbed ranged from 0.23
`to 0.97.2 This variability in plasma drug concen-
`tration may have accounted for the higher re-
`lapse rate seen in one study in which children
`with ALL treated with oral methotrexate were
`compared with those treated intramuscularly.'
`In addition, methotrexate absorption appears to
`be saturable, so that as the dose is increased, the
`fraction absorbed declines."9 Therefore, in pa-
`tients with low plasma drug levels, simply in-
`creasing the dose may not overcome poor bio-
`availability.
`The intramuscular route has been the prima-
`ry alternative to oral methotrexate in leukemia
`studies. Intramuscular methotrexate is rapidly
`and completely absorbed resulting in higher se-
`rum drug concentrations than following oral
`methotrexate. 7,9,'0 However, intramuscular in-
`jections must be administered by a health pro-
`fessional and may be difficult in chronically ill
`children with minimal muscle mass. In the pres-
`ent study, the subcutaneous route was evaluated
`as a parenteral alternative to oral administra-
`tion in children with ALL. The potential advan-
`tages for this route of administration include
`slow release of drug resulting in more prolonged
`exposure to methotrexate (a critical determi-
`
`1882
`
`Journal of Clinical Oncology, Vol 6, No 12 (December), 1988: pp 1882-1886
`
`Information downloaded from jco.ascopubs.org and provided by at Reprints Desk on June 1, 2016 from 216.185.156.28
`Copyright © 1988 American Society of Clinical Oncology. All rights reserved.
`
`Medac Exhibit 2015
`Frontier Therapeutics v. Medac
`IPR2016-00649
`Page 00001
`
`

`
`SUBCUTANEOUS METHOTREXATE PHARMACOKINETICS
`
`1883
`
`nant of cytotoxicity), ease of administration, and
`less variable, more complete absorption than
`that observed with oral administration. The fea-
`sibility of this approach was first studied in Rhe-
`sus monkeys and then in children with ALL at
`two different dose levels.
`
`MATERIALS AND METHODS
`
`Drug
`Methotrexate was obtained from commercial sources (Le-
`derle, Pearl River, NY). The standard intravenous (IV)
`preparation was used for both the IV doses and subcutane-
`ous doses in the animals and patients, and for the oral doses
`in monkeys. Standard 2.5 mg tablets were used in patients
`studied with an oral dose.
`
`Animals
`Five adult male Rhesus monkeys (Maccaca mulatta) rang-
`ing in weight from 4.8 to 10.1 kg (median, 8.8 kg) were
`studied. The animals were housed individually and received
`water and food ad libitum (animals were fasted overnight
`before the oral dose). Each animal was treated with metho-
`trexate at a dose of 1 mg/kg by three routes, orally, subcuta-
`neously, and by IV bolus, with the order of administration
`determined randomly. In addition, three of the animals re-
`ceived a 60-minute infusion of 1 mg/kg of methotrexate.
`Animals were given a minimum of 2 weeks to recover before
`the next dose was administered. Blood samples were drawn
`from a saphenous or femoral venous catheter, contralateral
`to the site of injection in the case of the IV doses. The
`heparinized specimens were obtained before the dose and 5,
`15, 30, 45, 60, and 90 minutes and 2, 3, 4, 6, 8, and 12 hours
`after the dose. Plasma was separated immediately by cen-
`trifugation and frozen at -20 0 C until assayed.
`
`Patients
`Twelve children (nine males and three females) with ALL
`in remission being treated on either an initial protocol (n =
`6, National Cancer Institute [NCI], Children's Hospital Na-
`tional Medical Center) or a relapse protocol (n = 6, St Jude
`Children's Research Hospital) participated in this study.
`The patients ranged in age from 3 to 19 years (median, 8
`years). Informed consent was obtained from the patients
`and their legal guardians before entry onto the study. All
`patients were in their first or second remission and receiving
`methotrexate as maintenance therapy on one of two sched-
`ules, either 7.5 mg/m2 orally, twice a week (six patients) or 40
`mg/m2 orally, once a week (six patients). The six patients at
`the low-dose level (actual mean dose received, 6.5 mg/m2)
`were studied following both their standard oral dose and an
`equal dose administered subcutaneously. At the high-dose
`level, four patients were monitored after oral, subcutane-
`ous, and IV bolus doses, one patient was studied after only
`an oral dose, and the sixth patient after only a subcutaneous
`dose. The order of administration was determined random-
`ly. Patients were fasted overnight before the oral dose. Com-
`plete blood counts, liver function tests, and renal function
`tests were routinely monitored on these patients before and
`
`1 week after each dose and demonstrated normal bone mar-
`row, hepatic, and renal function.
`Blood samples were collected in heparinized tubes before
`the dose and 15, 30, 60, and 90 minutes and 2, 3, 4, 6, and 8
`hours after the dose. Plasma was separated by centrifuga-
`tion and frozen at -20"C until assayed.
`
`Sample Analysis
`Methotrexate was measured with the dihydrofolate re-
`ductase inhibition assay which is specific for methotrexate
`.mo1/L.1l
`and has a lower limit of sensitivity of .001
`
`Pharmacokinetic Calculations
`Area under the plasma concentration-time curve (AUC)
`was derived using the linear trapezoidal rule and extrapolat-
`ed to infinity using the elimination rate constant derived
`from nonlinear regression analysis of the data.12 For the IV
`bolus doses the methotrexate concentration at time 0 used
`in the calculation of the AUC was the sum of the intercepts
`(A + B) derived from fitting the data to the biexponential
`equation below (using MLAB13 ):
`C(t) = Ae-at + Be-
`Absolute bioavailability (F) at the 40 mg/m 2 dose was
`calculated from the AUC using the following equation:
`
`AUCPO or sc . Dosei"
`AUCIV DosePO or SC
`(Abbreviations: PO, oral; SC, subcutaneous)
`
`Clearance was calculated by dividing the dose by the AUC.
`
`RESULTS
`
`Animal Study
`The plasma disappearance curves for metho-
`trexate administered subcutaneously, by IV bo-
`lus, and by 60-minute IV infusion are shown in
`Fig 1. The plasma methotrexate concentration
`following the subcutaneous dose peaked 15 to
`30 minutes after the dose, and ranged from 1.0
`to 2.3 pLmol/L. The sustained-release effect
`from subcutaneous administration can be ap-
`preciated from the curves. The plasma metho-
`remained above 0.1
`trexate concentration
`jLmol/L two- to three-fold longer with subcuta-
`neous administration than with IV bolus or infu-
`sion doses. Plasma concentrations following the
`oral dose are not shown because the absorption
`of methotrexate in these animals was poor
`(< 2% of the dose) and was not felt to be a
`representative model for humans.
`Table 1 lists the pharmacokinetic parameters
`for methotrexate administered by the various
`routes. The AUC for the subcutaneous dose ac-
`tually exceeded that for the IV bolus dose in all
`five animals studied. Since the dose was identi-
`
`Information downloaded from jco.ascopubs.org and provided by at Reprints Desk on June 1, 2016 from 216.185.156.28
`Copyright © 1988 American Society of Clinical Oncology. All rights reserved.
`
`Page 00002
`
`

`
`1884
`
`BALIS ET AL
`
`three animals received a repeat dose of metho-
`trexate infused IV over one hour to more closely
`simulate the levels achieved with the subcutane-
`ous dose. In these three animals the bioavail-
`ability of the subcutaneous dose was 102% ±
`12% with the 60-minute infusion is used as the
`standard. There was no local or systemic toxicity
`associated with the subcutaneous injection of
`methotrexate in Rhesus monkeys.
`
`Patient Study
`Figure 2 shows the plasma concentration-
`time profiles for subcutaneous methotrexate
`compared with oral administration at the 7.5
`mg/m 2 dose level (Fig 2A) and compared with
`IV and oral administration at the 40 mg/m2 dose
`level (Fig 2B). At the lower dose level subcuta-
`neous administration approximates the levels
`achieved with the oral dose. However, at the
`higher dose subcutaneous injection results in
`considerably higher plasma drug concentra-
`tions. Peak methotrexate concentration and
`AUC with the subcutaneous dose were four-
`than
`respectively,
`three-fold higher,
`and
`achieved with an identical oral dose (Table 2),
`and the subcutaneous dose provided exposure
`to _> 1 I.mol/L concentrations of methotrexate
`for up to six hours compared with 2.6 hours with
`the oral dose. The sustained-release effect with
`subcutaneous administration observed in the
`animals was not as evident in these patients
`when compared with the IV dose. Subcutaneous
`methotrexate was rapidly absorbed with the
`
`10
`
`1.0
`
`0.1
`
`Xu
`
`I-*
`
`a.
`
`0.01
`
`LU
`5l
`-
`
`0 I
`
`0.001
`
`0
`
`2
`
`4
`
`6
`
`8
`
`10
`
`12
`
`TIME [h]
`asma disappearance curves for metho-
`Fig 1. Pla
`hesus monkeys following administra-
`trexate in RI
`kg by IV bolus (0) (n = 5), 60-minute IV
`tion of 1 mg/I
`(A) (n = 3), and subcutaneously (0) (n
`infusion (Inf)
`nd error bars represent the geometric
`= 5). Points c
`me SD. Plasma concentration of metho-
`mean and on
`wing the subcutaneous dose is main-
`trexate follow
`e 0.1 pmol/L for 6.4 hours compared
`tained above
`3.3 hours for the IV bolus and Infusion
`with 2.2 and
`doses.
`
`cal for both
`due to the n
`dose or to
`methotrexat
`ting. As a
`
`routes, this difference may either be
`nore rapid clearance of the IV bolus
`an underestimation of the time 0
`te concentration from the curve fit-
`result of this unexpected finding,
`
`Table 1. Pharmacokinetic Parameters for Methotrexate
`Administered by Various Routes to Rhesus Monkeys
`Bioavailability
`of SC MTX
`(%)
`
`PO
`.089
`.153
`.056
`ND
`.044
`
`Total
`
`Clear-
`
`AUC
`Ani- Weight
`(pmol/L/h)
`IV Bolus
`IV Inf
`ancet
`SC
`mal
`(kg)
`IV Bolus
`IV Inf
`Std*
`Std*
`(mL/min)
`681P
`8.5
`2.88
`3.70
`4.08
`128
`91
`108
`687P
`8.8
`2.97
`3.52
`-
`-
`119
`107
`802F
`10.1
`4.73
`4.03
`4.12
`115
`117
`92
`631T
`4.8
`2.77
`2.52
`-
`110
`-
`70
`677J
`7.5
`4.65
`3.18
`4.60
`146
`101
`87
`Mean
`3.87
`3.00
`4.27
`124
`102
`93
`±SD
`0.83
`0.48
`0.28
`14
`12
`16
`Abbreviations: IV Inf, intravenous infusion; Std, standard; MTX, methotrexate; ND, not de-
`tectable.
`*Bioavailability calculated by dividing AUCSC by either AUCIv bolu or AUC4 inf.
`tClearance of the IV bolus dose.
`
`Information downloaded from jco.ascopubs.org and provided by at Reprints Desk on June 1, 2016 from 216.185.156.28
`Copyright © 1988 American Society of Clinical Oncology. All rights reserved.
`
`Page 00003
`
`

`
`SUBCUTANEOUS METHOTREXATE PHARMACOKINETICS
`
`1885
`
`10
`
`5 .
`..
`
`1.0
`
`0.1-
`
`l
`g---
`0
`
`0.001
`
`I-0
`w 0.01
`
`0.001
`
`TIME [h]
`
`TIME [h]
`
`Fig 2. Plasma
`disappear-
`ance curves for methotrexate in
`children with ALL following ad-
`ministration of (A) 7.5 mg/M 2
`orally (A) and subcutaneously
`(0)
`to six patients; and (B) 40
`mg/m 2 IV (0) (n = 4), orally (A) (n
`= 5), and subcutaneously (E) (n
`= 5).
`
`peak concentration occurring between 15 and
`30 minutes.
`Pertinent pharmacokinetic parameters are
`listed in Table 2. The peak plasma concentra-
`tion and bioavailability at the lower dose are
`equivalent for oral and subcutaneous metho-
`trexate, but the clear advantage for the subcuta-
`neous dose at the 40 mg/m2 level can again be
`appreciated. The subcutaneous dose was com-
`pletely absorbed, whereas, only 42% of the oral
`dose is bioavailable. The advantage for subcuta-
`neous administration can also be appreciated by
`comparing the relative bioavailability of the sub-
`cutaneous and oral 40 mg/m2 doses using the
`lower dose as a standard. The relative bioavaila-
`bility of the subcutaneous dose is 106% com-
`pared with 43% for the oral dose. When injected
`slowly, subcutaneous administration of metho-
`trexate was well tolerated in these children
`with ALL, with no evidence of local or systemic
`toxicity.
`
`DISCUSSION
`The results of this study indicate that the sub-
`cutaneous route appears to be a feasible route
`of administration for low-dose methotrexate in
`
`patients on an intermittent schedule (most
`maintenance regimens for ALL include weekly
`oral methotrexate at a dose of 15 to 20 mg/m2).
`Subcutaneous methotrexate was rapidly and
`completely absorbed at both dose levels studied,
`without evidence of local toxicity at the injection
`site.
`The finding of a greater AUC for methotrex-
`ate following subcutaneous administration com-
`pared with that resulting from an equal dose
`administered by IV bolus has two possible ex-
`planations. Total plasma clearance could be
`more rapid after the IV bolus. One could specu-
`late that renal tubular reabsorption is saturated
`at the initial high plasma methotrexate concen-
`trations or that the rapid injection does not al-
`low time for the complete tissue distribution
`and therefore less drug is available for slow re-
`lease at later time points. A more likely explana-
`tion is that the methods used to calculate the
`initial concentration (at time 0) underestimated
`that value leading to an underestimation of the
`AUC for the IV bolus dose.
`Figure 3 illustrates the advantage for subcuta-
`neous methotrexate over oral administration as
`the dose is increased. The AUC for a variety of
`
`Table 2. Pharmacokinetic Parameters for Methotrexate Administered
`by Various Routes and at Two Dose Levels to Children With ALL
`AUC
`Peak Conc.
`Bioavailability
`(%)
`(p.mol/L)
`(Lpmol/L/h)
`PO
`PO
`
`Dose
`(mg/m2)
`
`SC
`
`IV
`
`SC
`
`IV
`
`SC
`
`PO
`
`-
`
`7.5
`
`40
`
`2.99
`3.38*
`±1.65 ±1.34
`(n = 6)
`(n =6)
`22.0
`20.0
`7.93
`±+9.0
`t±3.5
`±2.88
`(n=5)
`(n=5)
`(n=4)
`*Data presented is the mean + one SD.
`
`0.94
`t+0.28
`(n = 6)
`7.40
`±3.00
`(n=5)
`
`0.98
`±0.46
`(n = 6)
`1.73
`t+0.35
`(n=5)
`
`-
`
`-
`
`-
`
`11.4
`+2.8
`(n=4)
`
`126
`42
`-t15
`±60
`(n=4) (n=4)
`
`Information downloaded from jco.ascopubs.org and provided by at Reprints Desk on June 1, 2016 from 216.185.156.28
`Copyright © 1988 American Society of Clinical Oncology. All rights reserved.
`
`Page 00004
`
`

`
`BALIS ET AL
`
`2
`
`sorption process. This saturation of absorption,
`which has been previously described for orally
`administered methotrexate, 2,6,9 was not ob-
`served with the subcutaneous dose.
`Because of its substantially better absorption
`at doses > 20 mg/m2, subcutaneous administra-
`tion appears to have a pharmacokinetic advan-
`tage over oral dosing in this dosage range. This
`bioavailability difference with increasing dose
`has also been recently observed in a study com-
`paring intramuscular and oral administration. 9
`In this study the intramuscular dose was com-
`pletely absorbed over a dosage range of 13 to 76
`.. ,hrnEc t1 h
`m,/-n
`nf t- h nral d-n ce
`nrnrnt
`f/
`absorbed fell from 42% at doses of _: 40 mg/m2
`to 18% at dose > 40 mg/m 2.
`Subcutaneous methotrexate may also be of
`use in selected patients on doses < 20 mg/m2
`who absorb the drug poorly from the gastroin-
`testinal (GI) tract. 14 Considering published
`data from one study on the superiority of paren-
`teral (IM) methotrexate in maintaining remis-
`sions in ALL,' a case could also be made for
`using intramuscular or subcutaneous metho-
`trexate during maintenance therapy in all chil-
`dren with ALL. Although not studied here, the
`subcutaneous route may also be useful for long-
`term, low-dose continuous infusion of metho-
`trexate.
`
`oral doses both from this study and several other
`published reports2,7,'14'15 is shown relative to the
`dose administered. At doses of oral methotrex-
`ate of 30 mg/m2 and above, the AUC plateaus,
`presumably as a result of saturation of the ab-
`
`REFERENCES
`1. Kearney PJ, Light PA, Preece A, et al: Unpredictable
`8. Smith DK, Omura GA, Ostroy F: Clinical pharmacol-
`serum levels after oral methotrexate in children with acute
`ogy of intermediate-dose oral methotrexate. Cancer Che-
`lymphoblastic leukaemia. Cancer Chemother Pharmacol
`mother Pharmacol 4:117-120, 1980
`3:117-120, 1979
`9. Teresi ME, Crom WR, Choi KE, et al: Methotrexate
`2. Balis FM, Savitch JL, Bleyer WA: Pharmacokinetics of
`bioavailability after oral and intramuscular administration
`oral methotrexate in children. Cancer Res 43:2342-2345,
`in children. J Pediatr 110:788-792, 1987
`1983
`10. Freeman-Narrod M, Gerstley BJ, Engstrom PF, et al:
`3. Pinkerton CR, Welshman SG, Kelly JG, et al: Pharma-
`Comparison of serum concentrations of methotrexate after
`cokinetics of low-dose methotrexate in children receiving
`various routes of administration. Cancer 36:1619-1624, 1975
`therapy for acute lymphoblastic leukemia.
`maintenance
`11. Falk LC, Clark DR, Kalman SM, et al: Enzymatic
`Cancer Chemother Pharmacol 10:36-39, 1982
`assay for methotrexate in serum and cerebrospinal fluid.
`4. Pinkerton CR, Welshman SG, Bridges JM: Serum pro-
`Clin Chem 22:785-788, 1976
`files of methotrexate after its administration in children with
`acute lymphoblastic leukemia. Br J Cancer 45:300-303, 1982
`12. Gibaldi M, Perrier D: Pharmacokinetics (ed 2). New
`5. The Medical Research Council's Working Party on
`York, Dekker, 1982, pp 445-449
`Leukaemia in Childhood: Medical Research Council leu-
`13. Knott GD: MLAB-A mathematical modeling tool.
`kaemia trial, UKALL VII. Arch Dis Child 60:1050-1054,
`Comput Programs Biomed 10:261-280, 1979
`1985
`14. Stuart JFB, Calman KC, Watters J, et al: Bioavailabil-
`6. Henderson ES, Adamson RH, Oliverio VT: The meta-
`ity of methotrexate: Implications for clinical use. Cancer
`bolic fate of tritiated methotrexate II: Absorption and excre-
`Chemother Pharmacol 3:239-241, 1979
`tion in man. Cancer Res 25:1018-1024, 1965
`15. Steele WH, Stuart JFB, Lawrence JR, et al: Enhance-
`7. Campbell MA, Perrier DG, Dorr RT, et al: Methotrex-
`ment of methotrexate absorption by subdivision of dose.
`ate: Bioavailability and pharmacokinetics. Cancer Treat
`Cancer Chemother Pharmacol 3:235-237, 1979
`Rep 69:833-838, 1985
`
`Information downloaded from jco.ascopubs.org and provided by at Reprints Desk on June 1, 2016 from 216.185.156.28
`Copyright © 1988 American Society of Clinical Oncology. All rights reserved.
`
`1886
`
`24
`
`20
`
`V 16
`
`12
`
`j
`0
`
`4 8
`
`4 0
`
`50
`
`60
`
`0
`
`10
`
`40
`20
`30
`DOSE [mg/m 2]
`Fig 3. Relationship between the mean AUC and
`subcutaneous dose administration (0) at low and
`high levels compared with oral administration (0)
`at various dose levels. Data from the present study
`and others"s are included. The number next to each
`point is the number of data points averaged to yield
`the point. The oral dose AUC plateau indicates satu-
`ration of the absorptive mechanism.
`
`Page 00005