CENTER FOR DRUG EVALUATION AND RESEARCH
`
`APPROVAL PACKAGE FOR:
`
`APPLICATION NUMBER
`
`21-462
`
`Clinical Pharmacology and Biopharmaceutics
`Review
`
`

`

`CLINICAL PHARMACOLOGY AND BIOPHARMACEUTICS REVIEW
`
`NDA 21-462 AMENDMENT
`
`Drug:
`
`ALIMTA
`
`Generic name:
`
`Formulation:
`
`Indications:
`
`Applicant:
`
`Pemetrexed; N-[4-[2—(2-amino-3,4—dihydro-4-oxo-7H—
`pyrrolol[2,3-d1pyrimidin-5—yl)ethyl]-benzoyl]-L-glutamic
`acid; MTA; LY231514;
`
`solution of 200 or 1000 mg for intravenous
`7
`.._
`40 ml
`infusion; lyophilized powder of 20, 100 and 500 mg for
`reconstitution and intravenous infusion.
`
`ALIMTA plus cisplatin for the treatment of patients with
`' malignant pleural mesothelioma.
`
`Eli Lilly and Company
`Indianapolis, IN 46285
`
`OCPB Division:
`
`Division of Pharmaceutical Evaluation 1 (HFD-860)
`
`n >OND Division:
`
`Division of Oncology Drug Products (HFD-ISO)
`
`Submission Dates:
`
`10/24/02; 12/06/02; 03/24/03; 9/4/03; 9/9/03; 9/16/03;
`10/7/03
`
`EDR e—Files:
`
`See Appendix for EDR files
`
`Primary/Pharmacometric Reviewer: Brian Booth, PhD.
`
`Pharmacometrics: Roshni Ramchandani, Ph.D.; Atul Bhattram, Ph.D.
`
`Pharmacometric
`
`Team Leader:
`
`Joga Gobburu, PhD.
`
`Team Leader:
`
`N.A.M. Atiqur Rahman, Ph.D.
`
`Type of Submission:
`
`NDA-New Molecular Entity
`
`1. Executive Summary
`
`This amendment incorporates the correction of a typographical error in the
`“Recommendations” section of the original review. The applicant is seeking marketing
`approval for Alimta in combination with cisplatin for the treatment of patients with
`
`

`

`unresectable malignant mesothelioma (or patients who are not surgical candidates for
`malignant mesothelioma). The applicant demonstrated a survival advantage with Alimta
`and cisplatin in comparison to cisplatin alone.
`
`Recommendation: The Alimta NDA 21-462 is acceptable to the Office of Clinical
`Pharmacology and Biopharmaceutics.
`
`Phase 4 Commitments:
`
`No phase 4 commitments are recommended.
`
`’/5/
`
`/S/
`
`'
`
`3
`
`
`l—{eviewerz Brian Booth, PhD.
`Team Leader: NAM Atiqur Rahman, PhD.
`
`CC:
`
`NDA 21—462
`
`A
`HFD-lSO/Division File
`HFD-ISO/PGarvey, MHazarika, Bwhite, Johnson]
`HFD-860/MehtaM, Csahajwalla, RahmanNAM, Gobburul, BoothB
`CDR/Biopharm
`
`

`

`This is a representation of an electronic record that was signed electronically and
`this page is the manifestation of the electronic signature.
`
`Brian Booth
`
`12/23/03 01:30:28 PM
`‘BIOPHARMACEUTICS
`
`Mehul Mehta
`
`12/23/03 02:20:36 PM
`BIOPHARMACEUTICS
`
`

`

`II. Table of Contents
`
`1. Executive Summary
`A. Recommendations
`B. Phase 4 Commitments
`
`~
`
`' II. Table of Contents
`
`111. Summary of Clinical Pharmacology and Biopharmaceutics Findings
`
`_
`
`IV. Question Based Review
`_A. General Attributes
`
`B. General Clinical Pharmacology
`
`C. Intrinsic Factors
`
`‘
`
`D. Extrinsic Factors
`
`E. General Biopharmaceutics
`
`F. Analytical Section
`
`I.
`
`V. Detailed Labeling Recommendations
`
`V1. Appendices
`A. Proposed Package Insert
`
`B. Individual Study Reports
`
`.
`
`C. EDR files
`
`D. Consult Review
`
`E. Cover Sheet and OCPB Filing Review Form
`
`1
`2
`2
`
`3
`
`4
`
`5
`5
`
`6
`
`10
`
`13
`
`19
`
`19
`
`23
`
`26
`26
`
`44
`
`130
`
`131
`
`

`

`III. Summary of Clinical Pharmacology Findings
`
`Alimta is a novel antimetabolite that inhibits thymidylate synthase, dihydro folate
`reductase and glycinamide ribonucleotide formyl transferase, and mediates cell death by
`inhibiting DNA synthesis. Alimta plus cisplatin mediates a survival advantage compared
`to cisplatin alone in malignant mesothelioma. The main toxicity of Alimta is
`neutropenia, but leukopenia, thrombocytopenia, stomatitis, vomiting diarrhea and nausea
`were also noted. The pharmacokinetics of Alimta follow a 2-companment model, and
`excretion is predominantly renal. Alimta was not metabolized by any cytochrome P-450,
`nor did it inhibit any cytochrome P-450 isozyme. Total systemic clearance of Alimta is
`91.8 mI/min and is well correlated with glomerular filtration rate and creatinine clearance
`(CLcr) calculated using the Cockcroft-Gault formula. The elimination half-life is 3.5
`hours, and no accumulation was noted. The pharmacokinetics of Alimta were not
`affected by sex, age or ethnicity. Co-administration of cisplatin did not alter the
`pharmacokinetics of Alimta, or vice versa. Co-administration of carboplatin did not alter
`the phannacokinetics of Alimta, but the pharmacokinetics of carboplatin may have been
`affected. F olic acid/vitamin BIZ did not alter the pharmacokinetics of Alimta, nor did
`aspirin at doses of 1.3 mg/day. However, ibuprofen increased Alimta AUC by
`approximately 20% at a moderate dose of 1.6 gin/day. Moderate doses of aspirin (1.3
`rug/day) did not affect the pharmacokinetics of Alimta. AspisirnRenaI impairment studies
`V of Alimta as a single agent indicated that the Alimta AUC increased by 130% in patients
`with moderate renal impairment (CLcr 30-50 ml/min; n=6), suggesting that neutropenia
`might be exacerbated in these patients.
`
`

`

`lV. Question Based Review
`
`A. General Attributes
`
`What are the highlights ofthe chemicalproperties ofdrug substance andformulation
`of the drug product?
`‘
`
`Alimta (pemetrexed;MTA;LY231514) is antifolate antineoplastic agent similar to
`methotrexate. The chemical name for ALIMTA is N-[4-[2-amino~4,7-dihydor-4-oxo-1H-
`pyrrolo [2,3—d]pyrimidin-5-y1)ethyl]benzoyl]-L-g1utamic acid disodium salt, which has a
`molecular weight of 597.49. The chemical structure is shown in Figure 1.
`
`LY231514
`
`
`
`(rsn
`
`Figure 1. The chemical structure of Alimta
`
`_.
`
`. Alimta is available as a lyophilized powder containing 500 mg of Alimta and 500 mg of
`' mannitol. Vials are reconstituted with 20 m1 of 0.9% saline (USP) and pH may be
`adjusted with hydrochloric acid or sodium hydroxide.
`
`i What is the therapeutic indication andproposed dosage regimen ofA limta?
`
`.
`
`V
`
`Alimta in combination with cisplatin is indicated for the treatment of patients with
`malignant pleural mesothelioma whose disease is either unresectable or who are not
`candidates for curative surgery. The proposed dosing regimen is 500 mg/mZ of Alimta
`infused over ten minutes, which is followed 30 minutes later with an infusion of 75
`‘ mg/m2 cisplatin over, two hours once every 21 days. Treatment is to be preceded with 5
`daily oral doses of folic acid and one intramuscular vitamin B12 injection. Vitamin B12
`should be repeated once every three cycles. Folic acid should continue during the full
`- course oftherapy and for 21 days after the last dose of Alimta.
`
`What is the putative mechanism of action ofAlimta?
`
`

`

`Alimta inhibits thymidylate synthase (TS), which is essential for with DNA synthesis,
`and thereby mediates its cytotoxic activity. Alimta also inhibits dihydrofolate reductase
`(DHFR), and glycinamide ribonucleotide formyl transferase (GARFT) (See Figure 2),
`
`10, Formleetra
`/ HydroFolate
`
`dUMP
`
`5,10 Methylene
`TetraHydro
`Folate \.
`
`Purine
`Synthesis
`
`GARFT
`
`Thymidylate
`Synthase
`
`.
`
`‘
`
`TetraHydroFolate
`
`Inhibition
`
`Inhibition
`
`DiHydroFolate
`
`Alimta
`
`dTMP
`
`Inhibition
`
`Alimta
`
`
`DiHydroFolate
`Reductase
`
`+ NADP
`
`Alimta
`
`DNA Synthesis
`
`Figure 2. Mechanisms of action for Alimta.
`
`What effectiveness and safety information contributes to the assessment of the clinical
`pharmacologz data?
`
`The effectiveness of Alimta in the pivotal clinical trial was based on prolonging survival.
`In a study of 574 patients with malignant mesothelioma, 448 were evaluable for response.
`The secondary endpoint was response rate. Toxicity (adverse events) that were assessed
`for Alimita were'neutropenia, leukopenia, thrombocytopenia, stomatitis, diarrhea, nausea
`and vomiting.
`
`B. General Clinical Pharmacology
`
`What are the effectiveness and safety endpoints?
`
`The primary clinical endpoint in the pivotal trial of Alimta plus cisplatin vs cisplatin
`alone was survival. Best overall responses were the secondary endpoint for this study.
`Responses were defined as complete response (CR), partial response (PR), stable disease
`(SD) or progressive disease (PD) based on tumor size assessed by CAT scan Two
`hundred and twenty six patients were treated with 500 mg/m2 of Alimta and 75 mg/m2 of
`
`

`

`cisplatin in the first cycle, compared to 222 who received 75 mg/m2 cisplatin alone.
`Thirty-six patients had their doses of Alimta reduced at various cycles.
`
`What are the active moieties in the plasma?
`
`Alimta itself is the active moiety. Alimta does not appear to be metabolized to any
`appreciable extent, and it is almost completely excreted intact in urine.
`
`What are the characteristics of the exposure-response relationships ofA limta?
`
`Effectiveness
`
`In the pivotal phase 3 trial, only one dose of Alimta was administered. Consequently, no
`concentration-effect relationships could be determined because of the narrow range of
`doses and AUCs available in this study. However, proportional hazards regressions
`indicated that survival was significantly correlated with Alimta treatment, especially if
`the patients were supplemented with folic acid/vitamin B12. Furthermore, complete and
`partial responses and stable disease were also correlated with Alimta use.
`
`Toxicity
`
`Neutropenia
`The main toxicity associated with Alimta was hematological. Neutropenia was the most
`significant toxicity, and was the main cause of dose reductions in the trial (9 of 36 cases).
`Alimta was significantly correlated with neutropenia leukopenia, thrombocytopenia,
`vomiting, and diarrhea.
`
`The applicant modeled neutropenia in response to single-agent Alimta (based on data
`from eight phase 2 studies). Simulations with this model suggested that neutropcnia will
`
`F'nal Model (paw): area on Nadir Depth
`
`1
`
`r-——r——1———r——r—-—r——r-"'fi
`D
`60
`100
`15D
`20)
`250
`320
`393
`
`LY231514 AUG ([1 g~hr3mL)
`
`& 0
`
`.!
`
`a”
`
`O
`
`ModelPrtdidedHANG(xum;
`
`
`FRED)
`
`Figure 3a. Simulations of nadir ANC in patients treated with single -agent Alimta.
`(from applicant)
`
`

`

`increase as AUC increases. These predictions are relatively consistent with the clinical
`findings, however, the predictive ability of this model is limited by the use of single-
`agcnt Alimta, and patients with relatively high renal function (>50 ml/min, mild
`impairment to normal). As cisplatin itself causes neutropenia, and renal function is
`related to Alimta AUC, it would be expected that neutropenia would be more severe with
`Alimta and cisplatin co-administration.
`
`Renal toxicity
`
`The data demonstrate a gradual decrease in CLcr over time with Alimta treatment in the
`renal impairment study (JMAW), but was not observed in the pivotal clinical trial
`(JMCH), deSpite patients being co-treated with cisplatin, which itself is renal toxic. The
`reason for this discrepancy may be the steps prescribed in the JMCH protocol to adjust
`doses or delay doses based on toxicity; Further, there was only one patient in study
`JMCH whose Alimta dose was reduced due to a reduction in creatinine clearance. A
`
`potentially confounding issue is that the patients predominantly had very high renal
`function; therefore, small changes in creatinine clearance that resulted in a lower value
`within the defined normal range may not have prompted dose modification.
`
`APPEARS THiS WAY
`ON ORiGliiii-‘sl
`
`APPEARS HHS WAY
`0W Oillfilfifil.
`
`

`

`The pharmacokinetics of Alimta were linear up to dosages of 700 mg/mZ.
`
`run
`
`I
`0
`V
`v-
`a
`I
`O
`o
`.
`
`mo
`
`3
`Em)
`u:
`.=
`aim)
`é
`3300
`
`i
`521:)
`a.J
`
`
`1m mmmmtmtantmtflom
`
`
`am
`
`m
`
`an
`
`g
`3
`g...
`:3ID
`F'10::
`S
`a!
`
`2m
`
`«10
`
`an
`
`um um «on
`mu
`Dose (mg)
`
`um um
`
`‘
`
`Wm“
`
`Figure 4. Linearity of Alimta AUC and Cmax (from applicant)
`
`The pharmacokinetics were described by a two-compartment model. Clearance is 91.8
`ml/min, volume of distribution is 16.1 L and the elimination half-life of Alimta is 3.5
`hours (see figure 5).
`
`mmmmammWET"
`
`Figure 5 Plasma concentration-vs-time curve for Alimta (from applicant)
`
`

`

`Excretion of Alimta was predominantly renal, and completed within 24 hrs.
`
`§
`
`
`
`i
`
`WmmwmwhwmHH'521S:388
`
`..9
`
`1‘!
`m J
`0:
`
`4
`
`1B
`12
`E
`mm mam man. hours
`
`2B
`
`31-
`
`Figure 6. Urinary accumulation of Alimta (from applicant)
`
`Clearance versus GFR
`
`Clearance m can,
`
`as
`
`m
`
`"
`g'”
`E
`”59)
`all!
`3
`:5“
`'43
`‘3
`£3
`
`a
`
`E
`
`m
`
`
`
`j?
`gm
`E
`if”
`:11!{
`gm
`a
`=1.
`R
`E
`
`u
`
`D
`
`__
`_'_f
`«vacuum—131nm
`Wmmwmp
`
`I
`
`ouanawmsnvnmm
`mmcmwwm
`
`Figure 7. Alimta systemic clearance as a function of GFR and CLcr (from applicant)
`
`Clearance of Alimta was well-correlated with renal function. Protein binding was
`approximately 80 % over a range of_5 to 200 pg/ml (ADME14), and was unaffected by
`mild or moderate renal impairment (JMAW).
`
`10
`
`

`

`C. Intrinsic Factors
`
`Does renal impairment affect the pharmacokinetics andpharmacodynamics of
`ALIM TA ?
`'
`
`The sponsor conducted a renal impairment study in patients with advanced cancer
`(JMAW). ALIMTA was administered as a 500 or 600 mg/m2 infusion over 10 minutes.
`Glomerular filtration rate was measured in patient using Tc99m-DPTA, and creatinine
`clearance was calculated using the CoCkcroft-Gault equation (CLcr) and CLcr using lean
`body mass (CLBM). Dense sampling was conducted over 72 hours in the first cycle of
`therapy. Data from forty-seven patients was used in the analysis. The sponsor
`demonstrated that the CLcr provided a good approximation of measured renal function
`(GFR) as shown in Figure 8.
`
`200
`
`50
`
`.A 0!O
`
`
`
`CLcr,mllmin aG
`
`o
`
`20
`
`4o
`
`60
`
`so
`
`100
`
`120
`
`140
`
`160
`
`GFR, mllmin
`
`' Figure 8. The relationship of creatinine clearance (CLcr) versus measured renal
`function (GFR; FDA analysis).
`
`I]
`
`

`

`According to the sponsor’s stratification, (upper normal >80 ml/min, lower normal 60-79
`mI/min, moderate 40-59), a 5 to 9-fold difference in ALIMTA CL can be expected
`between normal and severe impairment. Folic acid and vitamin BIZ had no apparent
`effect on ALIMTA pharmacokinetics. Despite the range in CL, the sponsor reported that
`ALIMTA was well tolerated, and concluded that no dose modifications were necessary in
`these patients with moderate to normal renal function. The only patient with severe renal
`impairment (CLcr 19 ml/min) died from drug related toxicity.
`
`These analyses were conducted with patients who were stratified by GFR. However, in
`practice, patients will be stratified by CLcr. Therefore. these data were re-analyzed
`accordingly, and stratified according to the renal impairment criteria promulgated by the
`FDA guidance. According to this analysis, ALIMTA AUC varied with renal function
`according to a curvilinear relationship (Figure 9).
`
`
`
`Dose-NormalizedAUC“.fp.os:e.-J-‘n0')onON
`
`P N
`
`O
`
`J
`
`0.0 r———_f—_r———,—_——_—fi———_—_,———_
`0
`so
`100
`150
`200
`
`CLcr, ml/min
`
`Figure 9. ALIMTA AUCinf in patients with varying renal function (CLcr; FDA
`analysis).
`V
`
`The ALIMTA Cmax and AUCinf for patients with different renal function are listed in
`table 1.
`
`TABLE 1. ALIMTA Cmax and AUC in patients with Renal Impairment (FDA
`Analysis)
`'
`
`12
`
`

`

`
`
`
`
`
`dose-normalized
`
`dose-normalized
`
`chane
`
`> 80 ml/min
`
`
`
`50-80 ml/min
`
`30-50 ml/min
`
`Severe
`<30 ml/min
`
`'
`
`
`
`
`
`1.182
`.
`
`512 T
`
`*Patient died from drug-related toxicity
`
`Using the CLcr and FDA criteria, the ALIMTA AUCin; increased more than 40% in
`patients with mild impairment, and 132 % in patients with moderate impairment. The
`variability is relatively constant (20-30%) across the groups, suggesting that the mean
`AUCinf estimates are likely fairly accurate. However, there are only six patients with
`moderate renal impairment, which is likely too few to adequately assess toxicity.
`Two additional issues need to be considered. First ALIMTA is indicated for use in
`
`combinationwith cisplatin, which itself induces renal toxicity and could reasonably be
`expected to exacerbate any ALIMTA toxicity in patients with renal impairment. This
`possibility was not addressed in these studies and the need for dose redUctions in patients
`with renal impairment should be addressed. Secondly, the effect of this combination
`should be addressed in multiple cycles of therapy because cisplatin toxicity is generally
`manifested after several cycles of therapy. Moreover, the mean renal function in these
`patients appeared to decline with each visit (see figure 10). Although this decline in renal
`function may have resulted from patients’ diseases, it may also have resulted from
`ALIMTA.
`
`
`
`MeanCLcrml/min
`
`105
`
`100
`
`90
`
`10GI
`
`o
`
`2
`
`4
`
`6
`
`8
`
`1o
`
`‘ Visit
`Figure 10. Mean Renal Function per visit (FDA analysis).
`
`13
`
`

`

`A similar phenomenon was observed for some patients treated with ALlMTA and
`Cisplatin (see figure 11).
`
`id:6021.00
`
`
`
`
`
`160
`
`80
`
`aU)U
`
`‘ 160
`
`80
`
`
`
`
`
`
`160
`
`80
`
`
`
`
`
`3-
`
`8
`
`13
`
`-
`
`. 18
`visit
`
`Figure 11. Renal function by visit from several patients in study JMCH (FDA
`' analysis).
`
`Gender did not appear to have any significant effect on ALIMTA exposure.
`
`D. Extrinsic Factors
`
`Is M TA involved in any significant Cycotchrome P-450 based drug-drug interactions?
`
`MTA does not appear to undergo metabolism to any significant extent, and plasma
`concentrations of ALMITA are not subject to modulation by inhibitors or inducers of
`cytochrome P-450 (CYP 450). Furthermore, in vitro studies indicate that ALIMTA does
`not significantly inhibit CYP 3A4, 2D6, 2C9 or 1A2. Therefore, ALIMTA is not
`expected to affect the disposition of co-administered medications that are metabolized by
`CYP 450 via enzyme inhibition. The potential of ALIMTA to induce CYP 450 isozymes
`
`14
`
`

`

`has not been determined, therefore it remains unknown whether MTA might cause a
`drug-drug interaction based on CYP 450 enzyme induction.
`
`Is MTA metabolized?
`
`_No. MTA does not appear to undergo significant metabolism. Preclinical mass balance
`studies indicated that the greatest amount of the radio-labeled species excreted in the
`urine was unchanged parent (>90% of the administered dose; ADME report 15). The
`only metabolite identified in preclinical studies, LY368942, was not detected in humans.
`In humans, unchanged MTA was the majorspecies detected in urine. A metabolite,
`LY338979 accounted for less than 5% of the species detected (based on ion intensity),
`and there were trace quantities of several other metabolites that could not be identified
`(ADME report 15).
`
`Does MTA inhibit cytochrome P-450 isozymes?
`
`The sponsor tested whether MTA inhibits cytochrome P-450 (CYP450) isozymes 3A,
`3D6, 2C9 and 1A2, by incubating MTA with microsomes that specifically expressed each
`CYP 450 isozyme. Microsomes were incubated with MTA in concentrations up to 1000
`uM (5.79 ug/ml). CYP 3A was inhibited by approximately 20% at an MTA
`concentration of 530 ng/ml. However, this is likely not significant because the MTA
`Cmax following the prescribed 500 mg/m2 dose of MTA is approximately 100 ng/ml.
`CYP-s 2D6, 2C9, 1A2 were not significantly affected by MTA at even higher
`concentrations (see Table 2).
`
`Table 2. In Vitro Inhibition of CYP 450 Isozvmes b MTA
`LY231514 Concentration
`CYP 450
`Metabolism
`
`
`
`
`
`
`
`
`
`
`
`»
`
`
`
`
`
`
`
`
`
`2c9
`
`000 (597)
`1000 597
`'as measured by midazolam metabolism to l'-hydroxymidazolam
`! as measured by bufurolol metabolism to l‘-hydroxybufurolol
`'as measured by diclofenac metabolism to 4’-hydroxydiclofenac
`@ as measured by phenacetin metabolism to acetominophen
`
`-m 1
`
`Does MTA induce cytochrome p-450 isozymes?
`
`This question remains unanswered because the applicant did not conduct any in Vitro
`studies to address this question. The applicant indicated that enzyme induction seems
`unlikely because 500 mg/m2 ALIMTA is administered as a 1‘0-minute infusion once
`every 21 days for mesothelioma, and enzyme induction generally requires a more
`prolonged incubation.
`
`Is M TA involved in any significant Non-CYP-450 based drug-drug interactions?
`
`15
`
`

`

`MTA is predominantly eliminated by renal excretion. Therefore, some of the co-
`administered medications'may be affected by MTA, and/or vice versa. The sponsor
`investigated the effect of co-administering MTA with cisplatin or carboplatin, and
`nonsteroidal anti-inflammatory drugs such as aspirin and ibuprofen, and the effect of
`simultaneously administered vitamins.
`
`Cisplatin
`
`The potential interactions between cisplatin and ALIMTA were evaluated in two studies.
`Study JMAP was a phase 1, dose escalation trial beginning with 500 mg/m2 of ALIMTA
`and 75 mg/m2 of cisplatin administered once every 21 days to 15 patients with advanced
`cancers (renal function ranged from53 to 104 ml/min). ALMITA infusions were 10
`minutes in duration, followed 30 minutes later by Cisplatin infusions (2 hrs). Patients
`were hydrated before and following the cisplatin infusion (1000 and 2000 m1 saline,
`respectively) (Treatment A). A second treatment arm was added in which ALIMTA and
`cisplatin infusions were separated by 24 hours (Treatment B). Dense sampling for
`ALIMTA was conducted over 24 hours in arm A, and over 97 hours in treatment arm B.
`
`Similar sampling was conducted for cisplatin. ALIMTA plasma concentrations and
`plasma total platinum were assayed. The results ofthese studies are listed in Tables 3
`and 4 below.
`
`TABLE 3. ALIMTA harmacokinetics with and without Cis . latin
`
`
`Parameter
`Treatment A ,
`Treatment B
`Treatment B
`
`
`600 m/m2 n=5
`97.1 21%
`
`
`
`
`
`
`
`500 m_/m2 n=6
`600 m_/m2 n=4
`'Cinu u/ml
`72.2 49%
`83.1 21%
`
`
`901 63%
`672 33%
`
`
`
`1 Parameter
`
`TreatmentA
`
`75 m/m2 n=4
`
`_
`
`7.12 29%
`
`Cm; u/ml
`CL ml/min/m2
`
`2.58 17%
`9.31 65%
`
`2.62 49%
`
`No significant interactions between ALIMTA and cisplatin could be observed in this
`study.
`
`The interaction between cisplatin and ALMITA was also studied as a part of the
`- phase 3 pivotal trial using a population pharmacokinetic approach. Mesothelioma
`patients were treated with 500 mg/m2 ofALIMTA and 75 mg/m2 of cisplatin
`administered once every 21 days. ALMITA infusions were 10 minutes in duration,
`followed 30 minutes later by Cisplatin infusions (2 hrs). The median clearance of
`_ ALIMTA was 88.4 ml/min (48.8 ml/rnin). Inclusion of cisplatin as a covariate to
`describe ALIMTA clearance had no significant effect. Volume of distribution, however,
`was reduced significantly by 3 %. The clinical significance ofthis finding is unclear.
`
`l6
`
`

`

`Similarly, the cisplatin clearance in this study was measured as 12.3 ml/min (total
`platinum). This clearance was also unaffected by the inclusion of ALIMTA as a
`covariate.
`
`Carboplatin
`
`Potential pharmacokinetic interactions between MTA and carboplatin were evaluated in a
`phase 1 study (IMUA) of 20 patients with pleural mesothelioma. This trial was a phase 1
`study of escalating doses of ALIMTA administered as a 10-minute infusion followed 30
`_minutes later by a 30-minute infusion of carboplatin. Both treatments were repeated
`every 21—days. Starting ALIMTA doses were 400 mg/mZ, and carboplatin was dosed to a
`target 'AUC of 4 mg/mlomin. Both ALIMTA and carboplatin were densely sampled (13-
`14 samples over 24 hours) in the first cycle. Total platinum in plasma was assayed for
`carboplatin, and the AUCOm for platinum ultrafiltrate was derived using the Ghazal-
`Aswad method (1996). There was no period when ALIMTA or carboplatin were
`administered alone. Results from these studies were compared to historical data. The
`pharrnacokinetics of ALIMTA were similar to parameter estimates observed from other
`phase ] studies which suggest that carboplatin does not alter ALIMTA in patients with
`normal renal function, at least in the first cycle of therapy.
`
`TABLE 5. ALIMTA CLEARANCE
`
`ALIMTA Dosao e CL ml/min/mZ
`
`
`
`The results for total carboplatin suggest that ALIMTA increases the clearance of
`carboplatin (see Table 6).
`
`TABLE 6. Carboplatin Pharmacokinetic Parameters
`
`
`
`The sponsor concluded that total carboplatin pharmacokinetics were not reliable because
`only two sampling times were available to assess terminal elimination. Instead they
`
`17
`
`

`

`calculated free platinum in the plasma ultrafiltrate as a function of the 24-hour total
`carboplatinum sample using the Ghazal-Aswad method. The resulting platinum
`
`ultrafiltrate AUC is 108 pg/mlomin is within the range of values reported in the Ghazal-
`
`Aswad report for patients dosed with 400 mg/m2 carboplatin ( “ ' ug/mlomin).
`The reliability ofthis approach is questionable, because it depends on the accuracy ofa
`single sample per patient instead of the entire data set. Furthermore, in the Ghazal-
`Aswad study only 5 patients were dosed with 400 mg/m2 carboplatin, therefore the
`validation of this approach is not satisfactory.
`Overall, the study appears to indicate that ALLMTA pharmacokinetics are not
`significantly affected by carboplatin, at least not during the first cycle of treatment.
`ALIMTA elimination may be affected by carboplatin in later cycles of therapy. These
`data suggest that ALIMTA increases the clearance of carboplatin, which could lead to
`underexposure of carboplatin. The mechanism of this interaction is unclear, and the
`interpretation of the study is hampered by lack of an appropriate in-study control.
`
`Non-Steroidal Anti-Inflammatory Drugs (NSAIDS)
`
`Aspirin
`
`Aspirin has a known effect on the pharmacokinetics of methtrexate, a congener of
`ALIMTA. A phase 1 study in 24 patients with advanced cancer was conducted to assess
`the effect of aspirin on the pharmacokinetics of ALIMTA. The patients also had varying
`degrees of renal impairment (IMAW(2b)). The mean creatinine clearance was 119
`ml/min (range: —-——
`‘ ml/min). The study was a two-way crossover design to
`facilitate comparison of ALIMTA PK in the presence and absence of aspirin. ALIMTA
`was administered as a 500 mg/m2 infusion over 10 minutes once every 21 days. Enteric-
`coated aspirin was administered as 325 mg every 6 hours 2 days before ALIMTA
`administration (13 g/day), and then one tablet was administered one hour prior to
`ALIMTA administration. Twelve samples for ALIMTA were obtained over 72 hours.
`Aspirin concentrations were not assessed. The results ofthe study are listed in Table 7.
`
`TABLE 7. ALIMTA Pharmacokinetics with/without As - irin
`
`
`
`
`
`
`
`
`
`
`
`
`
`These data suggest that there is no significant effect of aspirin on ALIMTA
`pharmacokinetics. However, the study does not assess a potential interaction with more
`prolonged exposure to aspirin, nor does it truly assess the effect that renal impairment
`may contribute as there were only three. patients with mild renal impairment (the
`remaining patients had normal renal function). Additionally, the dose used in this study
`(1.3 g/day) is moderate compared to the potential amount of aspirin that might be
`administered to these patients (2.5 to 3.9 g/day). These latter doses were the doses in
`which the methotrexate-aspirin interaction was observed. Therefore, interactions at
`
`18
`
`

`

`higher doses may be possible. Another shortcoming of this study is that the effect of
`ALIMTA on salicyclate excretion was not assessed. Renal excretion of salicyclate may
`be 30%, and it increases as the dose of aspirin increases. Therefore, the potential for
`ALIMTA to affect salicyclate excretion should also be addressed.
`
`Ibuprofen
`
`Ibuprofen has a known effect on the pharmacokinetics of methotrexate, a congener of
`ALIMTA. A phase 1 study in 24 patients with advanced cancer was conducted to assess
`the effect of ibuprofen on the pharmacokinetics of ALIMTA. The patients also had
`varying degrees of renal impairment (JMAW(2b)). The mean creatinine clearance was
`115 ml/min (range:
`. — ml/min). The study was a two-way crossover design to
`facilitate comparison of ALIMTA PK in the presence and absence of aspirin. ALIMTA
`was administered as a 500 mg/m2 infusion over 10 minutes once every 2] days.
`Ibuprofen was administered as 400 mg (2-200 mg tablets) every 6 hours for 2 days before
`ALIMTA administration (1.6 g/day), and then 400 mg was administered one hour prior to
`ALIMTA administration. Twelve samples for ALIMTA were obtained over 72 hours.
`Ibuprofen concentrations were not assessed. The results of the study are listed in Table 8.
`
`
`TABLE 8. ALIMTA Pharmacokinetics with/without Ibu rofen
`
`
`
`
`
`
`
`
`
`AUCo-m
`
`
`g (mvmin/mz)
`
`
`
`
`Parameter
`
`
`
`ALIMTA alone ALIMTA+ ibu n rofen
`
`
`121 i 27.9%
`
`These data indicate that there is a statistically significant increase in ALIMTA AUC
`(~20%) and Cmax (~15%), and a significant decrease in CL (~17%).
`In this case as well,
`the study does not assess a potential interaction with more prolonged exposure to
`ibuprofen, nor does it truly assess the effect that renal impairment may contribute as there
`were only two patients with mild renal impairment (the remaining patients had normal
`renal function). Additionally, the dose used in this study (1.6 g/day) is moderate
`compared to the potential amount of ibuprofen that might be administered to these
`patients. Previous studies have shown that ibuprofen dosesof 2.4 to 3.6 g/day induced a
`40% reduction in total and renal clearance of methotrexate (Tracy 1992). Therefore,
`interactions at higher doses may be possible.
`
`5-Fluorouracil (5-FU)
`
`The effect of ALIMTA on S-FU (and vice versa) was to be studied in a phase 1 study of
`this drug combination in patients with locally advanced or metastatic cancer (JMAR).
`However, the third amendment to the protocol removed the pharmacokinetic objective
`from the study, and data was collected for only 2 patients. Consequently, potential drug-
`drug interactions for this combination have not been examined.
`
`Vinorelbine
`
`l9
`
`

`

`The effect of ALIMTA on vinorelbine (and vice versa) was to be studied in a phase 3
`study of this drug combination in patients with locally advanced or metastatic non-small
`cell lung cancer (JMBQ). However, the study was terminated prematurely due to low
`enrollment, and data was collected for only 3 patients. Consequently, potential drug-drug
`interactions for this combination have not been examined.
`
`E. General Biopharmaceutics
`
`The BCS, bioavailability and bioequivalence issues relevant to per oral formulations do
`not apply to ALIMTA as it is a reconstituted lyophilized or aqueous solution for
`intravenous administration.
`
`F. Analytical Section
`
`The sponsor submitted method validation studies for MTA, cisplatin, carboplatin, and
`folic acid.
`
`MTA (ALIMTA)
`
`Several assays were submitted for MTA quantification. Overall, there were two basic
`assays, one based on LC/MS/MS and the second was —,'
`'. The .
`__
`assay
`consisted ofthree pans.
`
`Method ~ was the LC/MS/MS method that assayed MTA concentrations from A
`~— ) ng/ml in human plasma. No interfering peaks were observed at the retention
`time for MTA. The limit of quantification (LOQ) was ”ng/ml, and the accuracy and
`precision measurements were below . —3. The accuracy ofthe QC samples were also
`less than ~— Vo for the room temperature (up to 24 hours) and autosampler;
`"7
`stability tests. The precision ofthe low QC sample t ~ng/ml) following three freeze-
`thaw cycles was slightly high \--‘/o), but this is not likely problematic. The accuracy and
`precision of the high QC samples was less than — 4;.
`
`Method ~— l was the LC/MS/MS method that assayed MTA concentrations from
`w ’ ng/ml in human plasma. No interference peaks were observed at the
`. retention time for MTA. The limit of quantification (LOQ) was _ _— ig/ml, and the
`accuracy and precision measurements were below ~Vo. The accuracy of the QC samples-
`were also less than 7‘" /o for the room temperature and autosampler stability tests. The
`accuracy and precision of the low and high QC samples (
`'3‘
`ng/ml) following
`three freeze-thaw cycles was ‘ % or less.
`
`Method . ~— was the LC/MS/MS method that assayed MTA concentrations from
`_ _,
`ng/ml in human urine. No interference peaks were observed at the
`retention time for MTA. The limit of quantification (LOQ) was —— l ng/ml, and the
`accuracy and precision measurements were below—’ 6. The precision of the low QC
`sample t ”T l ng/ml) at 4 hours was slightly high ( - %) and the accuracy at 24 hours
`was —— V0 at room temperature. However, the remaining measurements of accuracy and
`
`20
`
`

`

`precision at 4 and 24 hours were less than “ . for the room temperature and autosampler
`stability tests. The accuracy and precision of the low and high QC samples (
`'5
`~— ng/ml) following three freeze-thaw cycles was ~— 6.
`
`Feb94 was 3 HPLC method with ultraviolet (UV) detection.
`——~
`Method
`Human plasma concentrations of MTA were measured over ranges of m
`
`The accuracy and precision for each
`o.N0 interference was detected at the retention time
`concentration range was less than
`of MTA or its internal standard. The limit of quantification was rig/ml, where accuracy
`and precision were less than f‘J/o. No freeze-thaw, room temperature or autosampler
`stability data was reported.
`
`-01Jul94 was 3 HPLC method with ultraviolet (UV) detection
`Method —’
`Human urine concentrations of MTA were measured over ranges of
`~
`The accuracy and precision for each concentration range was ‘ '4; or less
`Some interference was detected at the retention time of MTA The signal to noise ratio
`for the interference was equivalent to s ug/ml MTA. The sponsor set the limit of
`quantification as‘ 1g/ml, however, according to the Bioanalytical Method Validation
`Guidance, the LOQ shouldbe set at 5 x h