Invesrigational New Drugs 14: 325-335, l996.
`© 1996 Kluwer Academic Publishers. Printed in the Netherlands.
`
`325
`
`A Phase I clinical study of the antipurine antifolate lometrexol (DDATHF)
`given with oral folic acid
`
`Sudsawat Laohavinij,* Stephen R. Wedge‘, Micheal J. Lind, Nigel Bailey, Alison Humphreys,‘
`Madeleine Proctor, Fiona Chapman, Dorothy Simmons, Avril Oakley, Lesley Robson,
`Lyndsey Gumbrell, Gordon A. Taylor, Huw D. Thomas, Alan V. Boddy, David R. Newell and
`A. Hilary Calvert
`Cancer Research Unit, The Medical School, University ofNewcastle-upon-Tyne, Framlington Place,
`Newcastle—upon-Tyne, NE2 4HH, UK
`
`Key words: lometrexol, lometrexol-toxicity, lometrexol-clinical efficacy, lometrexol and folic acid, DDATHF
`
`Summary
`
`Lometrexol is an antifolate which inhibits glycinamide ribonucleotide formyltransferase (GARFT), an enzyme
`essential for de nova purine synthesis. Extensive ercperirnental and limited clinical data have shown that lonretrexol
`has activity against tumours which are refractory to other drugs, notably methotrexate. However, the initial clinical
`development of lometrexol was curtailed because of severe and cumulative antiproliferative toxicities.
`Preclinical murine studies demonstrated that the toxicity of lometrexol can be prevented by low dose folic acid
`administration, i.e. for 7 days prior to and 7 days following a single bolus dose. This observation prompted a
`Phwc I clinical study of lcrnctrcxcl given with fclic aid supplementation which has ccnfirrncd that the toxicity
`of lometrexol can be markedly reduced by folic acid supplementation. Thrombocytopenia and mucositis were the
`major toxicities. There was no clear relationship between clinical toxicity and the extent of plasma folate elevation.
`Associated studies demonstrated that lometrexol plasma pharmacokinetics were not altered by folic acid admin-
`istration indicating that supplementation is unlikely to reduce toxicity by enhancing lometrexol plasma clearance.
`The work described in this report has identified for the lint time a clinically meptalzle schedule for the
`administration of a GARFI‘ inhibitor. This information will facilitate the future evaluation of this class of compounds
`in cancer therapy.
`
`Introduction
`
`(5,10-dideaza-5,6,7,8-tetrahydrofolate-
`Lometrexol
`(6R)-DDATHF) is a new folate analogue which was
`synthesized in 1985 by Taylor and colleagues [1].
`Unlike methotrexate, lornetrexol does not inhibit dihy-
`drofolate reductase, but acts instead against glyci-
`namide ribonucleotide fonnyltransferase (GARFT), an
`enzyme essential for de novo purine synthesis [2]. Both
`in vitro and in vivo, lometrexol has been shown to have
`antitumor activity against murine and human tumour
`
`“‘ Supported by Eli Lilly and Company, Indianapolis. IN, USA.
`Financial support was also provided by the Nonh of England Cancer
`Research Campaign.
`
`cells [2-4], and on the basis of its preclinical activity
`was selected for clinical trial.
`
`In previous Phase I studies of lometrexol when giv-
`en alone,
`the total dose of lometrexol which could
`be safely given was found to be only 10-12 mg/mz
`per course [5-7]. In marked contrast, in mice, 600
`mymz/week was tolerated in chronic toxicity studies
`[8]. Furthermore, the onset of profound myelosuppres-
`sion and/or mucositis in most patients 6—8 weeks after
`lometrexol administration prevented administration of
`more than two courses of therapy in most studies. Thus,
`it has not been possible to perform Phase H studies to
`evaluate the potential efficacy of lometrexol. Howev-
`er, evidence of antitumor activity was observed in the
`Phase I clinical studies of lometrexol, in patients with
`
`Lilly Ex. 2031
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2031
`Sandoz v. Lilly IPR2016-00318
`
`

`
`326
`
`malignantfihrous histiocytoma [9], non—small cell lung
`cancer, breast cancer and colorectal carcinoma [6].
`Following the initial clinical evaluation of lome-
`trexol, further studies were performed in mice in an
`attempt to ameliorate the cumulative toxicity of lome-
`trexol and hence enable repeated courses of the drug
`to be given. These studies revealed that the therapeutic
`eflicacy and toxicity of lometrexol were highly depen-
`dent upon dietary folic acid intake [10, 11] and these
`preclinical data prompted the Phase I study of lome-
`trexol given with folic acid supplementation described
`here. The objectives of this clinical Phase I study were:
`(a) To evaluate the effect of folic acid on lometrexol
`pharmacodynamics, in order to determine whether
`folic acid improves tolerance of lometrexol.
`(b) To determine the toxicity of lometrexol in patients
`receiving multiple courses of the drug with folic
`acid supplementation.
`(c) To describe the pharmacokinetics of lometrexol in
`patients receiving folic acid supplementation.
`
`Patients and methods
`
`Patient eligibility
`
`From September 1991 to December 1995, 43 patients
`with a histologically confirmed diagnosis of malig-
`nant solid tumour, which was refractory to estab-
`lished therapies or for which no standard therapy exist-
`ed, were entered into this study. All patients had a
`predicted life expectancy of at least 12 weeks, and
`had recovered from the toxicity of previous treatment
`before entering onto the study. Specifically, patients
`were required to not have received previous anti-
`cancer therapy or other investigational drugs within
`at least 4 weeks (6 weeks if prior therapy included
`a nitrosourea, mitomycin C or extensive radiothera-
`py). Exclusion criteria included factors which could
`have interfered with lometrexol dispositionltoxicity or
`folic acid absorption, and comprised; (a) concomitant
`medication with allopurinol, probenecid, nephrotoxic
`agents, trimetrmprin-., anti-epilcptics, co-trirnoxazole
`or pyrimethamine,
`(b) extensive radiotherapy and
`(c) inflammatory ulcerative bowel disease, or malab-
`sorption syndrome. Concurrent treatment with other
`experimental drugs or other anticancer therapies was
`not allowed. Patients with clinical evidence or symp-
`toms suggestive of coronary artery disease or central
`nervous system disease were excluded. Patients with
`effusions and/or ascites were also not recruited.
`
`All patients were required to have adequate organ
`function prior to treatment, with marrow function char-
`acterised by a white blood cell count of at least 4 x
`109/1, neutrophil count at least 2 x 10°/I, haemoglobin
`level of at least 10 gldl, and platelet count of at least 100
`x 109/I. Adequate hepatic function was also required,
`as characterised by bilirubin levels of < 25 _umol/I,
`alkaline phosphatase 5 2.5 times upper limit of nor-
`mal, alanine transaminase (ALT/SGPT) g twice the
`upper limit of normal, prothrombin and partial throm-
`boplastin time within normal range. The creatinine
`level was required to be less than 12.0 umolll and the
`glomerular filtration rate (GFR) to be above 50 ml/min
`as measured by 5‘Cr~EDTA clearance.
`
`Study design
`
`Folic acid (Approved Prescription Services Ltd.,
`Leeds, U.K.) was given daily as a single 5 mg tablet
`for 7 days prior to and 7 days following lometrexol
`administration at 4 week intervals. Lometrexol (Lilly
`Research Centre, Erl Wood Manor, Surrey, UK.) was
`reconstituted in 0.9% (w/v) saline and administered as
`a rapid i.v. bolus over 30 seconds to one minute at a
`concentration of 1-10 mg/ml. Patients were admitted to
`the Department of Medical Oncology, Newcastle Gen-
`eral Hospital, to receive lometrexol and were observed
`for a further 24 hours following drug administration, to
`ensure that acute toxicity was not apparent. The follow-
`ing studies were performed weekly: physical exarnina-
`tion, toxicity and performance status assessment, and
`biochemical analysis. Full blood counts were measured
`twice a week. As part of the Phase i trial of lometrex-
`ol with folic acid it was important to demonsuate that
`plasma folate concentrations of patients were increased
`by folate supplementation and folate levels were mea-
`sured on course 1 prior to supplementation (day 7)
`and after 7 days of iclate adrriinisiratiun ‘out prior i.U
`lometrexol (day 0). Plasma folate concentrations were
`determined using a commercial folate binding assay
`(SimulTRAC-SNB. Becton Dickinson, Oxford, UK).
`The starting dose was 12 mg/m2 as this dose of
`lometrexol given mane had been well tolerated on the
`first course of therapy in previous Phase I studies,
`regardless of schedule [S-8]. Lometrexol was given as
`a single bolus injection every 4 weeks, with 5 mg/day
`oral folic acid administration 7 days prior to treatment
`with lometzexol and 7 days afterwards on each. course.
`Toxicities were evaluated according to World Health
`Organisation (WHO) criteria. If repeated courses at a
`given dose level were tolerated without toxicity greater
`
`Lilly Ex. 2031
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2031
`Sandoz v. Lilly IPR2016-00318
`
`

`
`than WHO grade II, doses were escalated according
`to the clinical judgement of the investigator with the
`approval of the Medicines Control Agency (London,
`UK). Doses were initially escalated to 16, 30 and
`45 mg/m2 every 4 weeks. Subsequently, the interval
`between each course was altered to every 3 weeks
`as the maximum tolerated dose (MTD) had not been
`reached at 45 mg/m2 every 4 weeks and because animal
`experiments indicated that more frequent adrninist:ra-
`tion of lometrexol increased antitumour activity (Eli
`Lilly Co, unpublished results of G. Grindey, person-
`al communication). Therefore lometrexol was subse-
`quently administered every 3 weeks and doses escalat-
`ed from 45 to 60, 78, 100, 130 mg/m2 and 170 mg/m2.
`No intra-patient dose escalation occurred
`Patients were evaluable for therapy-related anti-
`tumour activity if they had received two or more cours-
`es of therapy and disease measurements were recorded
`over at least an eight-week period from the first dose
`of therapy, with maintenance of a response for at least
`1 month. A complete remission (CR) was defined as
`the disappearance of all tumour as assessed by phys-
`ical examination and non-invasive investigations. A
`diminution by > 50% of the product of two diameters
`of a tumour was considered a partial remission (PR).
`Progression was indicated by the development of new
`lesions or an increase of 25% or more in the sum of the
`
`products of diameters of measurable lesions.
`
`Phannacokineric studies
`
`Lometrexol pharmacokinetics were determined in 24
`patients receiving foiic acid supplementation and iome-
`trexol at all dose levels except 170 mg/m2. The method-
`ology used and the results obtained for the dose range
`12-45 mg/mz has been published separately [12].
`
`Results
`
`Forty-three patients (23 females, 20 males) were
`recruited into the study who received a total of
`O0 am"-c-an
`'T'l-.4 n.-4+.-Aanuano rd-loo--at-la1v:oO:r\¢v at 01..
`/I V\I\Ilpl\-hJ- AILV ‘J!-VLIVIILAAIVIII %lli|.||l»V|vVllIlLlUO VA IAJV
`
`patients are shown in Table 1. The median age was
`54 years (range 30-72 years). Thirty-five patients
`received at least 2 courses and were assessable for
`
`response and toxicity. Eight patients received only
`one course because of disease progression and were
`only assessable for toxicity. One patient who received
`only one course of lometrexol at 45 mg/m2 every 4
`weeks was ineligible due to the concurrent adminis-
`
`Table 1. Patient characteristics
`
`Total number of patient:
`Number of cotnses administered
`Sex
`Females
`Males
`Ase (yrs)
`Median
`Range
`Performance status (WHO)
`
`—o
`
`Tumour types:
`Colorectal carcinoma
`Breast carcinoma
`Non-small cell lung carcinoma
`Malignant melanoma
`Ovarian cancer-
`Pnncreatie carcinoma
`Renal cell carcinoma
`Unknown primary carcinoma
`Peritonml carcinoma
`Adenoconical carcinoma
`Previous treatments
`l"sawvuu|.-an»::
`\.¢Ivbllllllidflyj
`
`Gremothenpy and radiotherapy
`
`327
`
`No.
`
`43
`99
`
`23
`20
`
`54
`30-72
`
`53
`
`-—:-.—v-—\al;.I'uJ'JIO\-O¥Id
`
`—\n19..
`
`tration of allopurinoi but was in any case assessed.
`The majority of patients had been pretreated, 31 of 43
`patients (72%) had previous chemotherapy and 12 of
`43 patients (28%) had also received prior radiotherapy.
`
`Myeiotoxiciry
`
`Thrombocytopenia was the major toxicity observed in
`patients receiving 2 30 mglmz lometrexol. Grade III-
`IV thrombocytopenia was observed in 7% (1/15), 14%
`(2/14), 7% (1/15), 26% (2/10) and 33% (3:9) of cours-
`es at doses of 30 and 45 mg/m2 every 4 weeks, and 60,
`100 and 130 mg/ml every 3 weeks, respectively (Table
`2). However, the patient with grade IV thrombocy-
`topaenia at 30 mglmz developed ascites and paralytic
`ileus during treatment with lcmetrexcl ‘.-.'.'.~.icl-. might
`have efiected the absorption of folic acid. One patient
`receiving lometrexol at 45 mg/m2 every 4 weeks,
`who developed grade IV thrombocytopenia. also had
`bowel obstruction after receiving the second course.
`The two patients who developed grade IV t_|n'or_rrbo-
`cytopenia (1 at 30 mg/m1 and 1 at 45 mg/m2 every
`4 weeks) required hospital admission for leucovorin
`rescue, packed red blood cells and platelet transfu-
`
`Lilly Ex. 2031
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2031
`Sandoz v. Lilly IPR2016-00318
`
`

`
`328
`
`Incidence of throrrbocytopenia in patients treated with
`Table 2.
`lometrexol given with folate supplementation
`
`Dose
`level
`(mg/in‘)
`
`l2 q 4wl:
`16 q 4wk
`30q4wk
`45 q4wk
`45 1] 3w]:
`60q3wk‘
`78q3wk‘
`rooq 3wk
`130-.;3-.-.o.».
`I70q 3w):
`
`Total
`['11-
`tientr
`
`Total
`cour-
`ses
`
`3
`4
`5
`s
`3
`6
`3
`5
`5
`1
`
`7
`1 1
`15
`14
`s
`15
`s
`to
`9
`2
`
`Number of courses with
`WHO toxicity gruh
`1
`11
`111
`
`—
`—
`—
`1(1)
`1(4)
`1(1)
`-
`—
`241.2)
`1(2)
`
`-
`—
`-
`—
`1(2)
`—
`-
`—
`—
`—
`
`-
`-
`-
`1(1)
`—
`1(2)
`-
`2(1,2)
`1(3)
`-
`
`o
`
`7
`1 1
`14
`11
`6
`13
`t
`s
`A.
`1
`
`xv
`
`—
`—
`1"(2"')
`1(2)
`—
`—
`—
`—
`2(1.2)
`_
`
`“Patient with ascites and paralytic ileus.
`‘The course number on toxicity developed.
`°0nepatient each at 60 and 78 mglmz received lometrexol for6 and
`4 courses, respectively, without experiencing thrombocytopenia.
`
`crane!
`
`couz cums
`
`Gown 4
`
`Eiilx‘ M ‘A
`‘.1
`;%::: V
`\/V
`
`-to 0
`
`20
`
`so
`
`so
`
`so
`
`too
`
`120
`
`Time ldaysl
`
`Figure l. Lometrexol-induced cyclical reduction in platelet count.
`in one patient receiving 4 courses of lometrexol at 30 mg/m3 every
`4 weeks. ‘l'he timing of lometrexol administration is represented by
`arrows.
`
`sions. Intravenous leucovorin (30 mg) was given every
`6 hours for 12 days (patient at 30 mglmz every 4 weeks)
`or 14 days (patient at 45 mg/m’ every 4 weeks) until
`myelosuppressior. resolvfl. Althoufi the majority of
`patients did not develop thrombocytopenia, a cyclical
`decrease in platelet counts following successive cours-
`es of Iometrexol was observed in all patients and one
`example of a patient treated at 30 mglmz is illustrated
`in Figure 1.
`Leucopenia was infrequent and mild, the greatest
`toxicity observed in this study was grade IV in one
`patient (1/9 courses) and grade III in another patient
`
`(l/9 courses) at 130 mg/m2 every 3 weeks. Overall,
`grade I—T_l!eI_1eopen_ia was observed on 27 of99 courses
`(27%); however, grade III—IV toxicity was seen on only
`4 of 99 courses (4%). No leucopenia greater than grade
`I was observed in patients receiving 12 or 16 mg/m2 of
`lometrexol (18 courses).
`Neutropenia was also infrequent and mild. No
`patient at 12 or 16 mg/m2 lometrexol developed neu-
`tropenia (18 courses). Overall, grade I-II neutropenia
`was observed in 11% of courses (11/99) and grade
`III neutropenia developed in 2 patients (2I99 courses),
`one at 45 mg/rnz every 4 weeks and one at 130 mglmz
`every 3 weeks. Grade IV neutropenia with fever requir-
`ing intravenous antibiotics occurred in only one of 8
`patients (l/l4 courses) treated at 45 mg/ml every 4
`weeks (afterthe second course of treatment). Grade IV
`neutropenia without fever was observed in one patient
`(1Il0 courses) and 2 patients (2/9 courses) treated at
`100 and 130 mglmz, respectively.
`in patients who
`Anaemia became prominent
`received lometrexol for more than 2 courses. One
`
`patient treated at 30 mg/m’, 2 patients at 45 mglm2
`every 4 weeks and 2 patients at 100 mg/ml every 3
`weeks (6/99 (6%) of the treatment cycles) developed
`grade III or IV anaemia. The three patients who devel-
`oped grade III anaemia, one at 30 mg/m2, one at 45
`mg/mi and one at 100 mglmz, also had thrombocy-
`topenia. The other patient at 45 mg/m2 who devel-
`oped grade III anaemia after the first course of treat-
`ment, which progressed to grade IV after receiving
`the second course, had upper gastrointestinal bleeding
`due to non-steroidal anti-inflammatory drug adminis-
`tration (diciofenac) and no thrombocytopenia. There-
`fore severe anaemia (grade III—IV) was observed on
`only 4% of courses (4/99), and mostly developed in
`thrombocytopenic patients. Mild anaemia (grade I-II)
`was observed on 46% of courses (46/99); however,
`frequent blood sarnpling for pharmac-o‘:ai'-ietic studies
`may have been partly responsible.
`
`Gastrointestinal toxicity
`
`Oral mucositis, manifested by diff-see eryt.":e.'r.a and
`small ulcers of the buccal mucosa, soft and hard palates
`and tongue, was observed in patients receiving 2
`30 mg/m’ lometrexol. Grade I to [I oral mucositis
`occurred in 26 of 99 courses (26%). One of 6 patients
`at 60 |_'!lg/_|'!I2 lornetrexol every 3 weeks and the only
`patient treated at 170 mglmz developed grade III toxi-
`city after 2 courses of treatment (2 of 99 courses (2%))
`(‘Table 3).
`
`Lilly Ex. 2031
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2031
`Sandoz v. Lilly IPR2016-00318
`
`

`
`Table 3. Incidence of gastrointestinal toxicity in patialrs treated
`with lometrexol given with folate supplementation
`
`Number of courses with toxicity
`Total
`Total
`Dose
`level
`pa-
`cour- Mueositis
`Diarrhoea
`(mg/tn’)
`tients
`ses
`H1‘
`Ill’
`I-11‘
`m‘
`
`
`12 q4wk
`16 q 4wk
`30 q Iiwit
`45 q 4&3wk
`60 q 3wk
`78 q 3M:
`I00 q3wk
`I30 q 3wI<
`I704; 3wk
`
`3
`4
`5
`I I
`6
`3
`5
`5
`l
`
`7
`l I
`i5
`22
`I5
`8
`I0
`9
`2
`
`o
`0
`i(i6)"'
`l0(45)
`[(6)
`4(50)
`300)
`707)
`l
`
`0
`0
`6
`0
`1(6)
`0
`0
`0
`I
`
`o
`108)
`5(33)
`1(4)
`l(6)
`0
`0
`3(33)
`o
`
`0
`0
`i(6)
`1(4)
`0
`0
`l(l0)
`l(l 1)
`0
`
`Total
`43
`99
`26(26)
`2(2)
`l2(l2)
`4(4)
`
`
`329
`
`and naproxen), which can cause renal dysfunction.
`According to the WHO classification of renal toxic-
`ity no patient in this study experienced renal toxicity,
`i.e. there were no increases in blood urea or creatinine
`
`to > 1.25 x upper limit of normal values.
`
`Other toxicities
`
`Two out of the six patients at 60 mg/m2 every 3 weeks
`developed grade I and II muscular and joint pains a
`few days after treatment. No neurologic, cardiac, pul-
`monary, hepatic, cutaneous toxicities or alopecia were
`observed.
`
`“WHO toxicity grade.
`‘Percentage of courses associated with toxicity.
`
`Efficacy of leucovorin in the reversal of lometrexol
`toxicity
`
`Diarrhoea was infrequent and mild (grade I to II
`on 12 of 99 courses (12%), Table 3); although grade
`III diarrhoea requiring intravenous fluid and folin-
`ic acid rescue did occur in 4 patients, one each at
`30 and 45 mgimz every 4 weeks, and 100 and I30
`mg/m2 every 3 weeks. However, the grade III diar-
`rhoea observed in the one patient at 30 mg/m2 after the
`second course of lometrexol may have been disease
`rather than lometrexol-related because this patient had
`paralytic ileus and usciies.
`Mild nausea and vomiting (grade I to II) was
`observed in 37 of 99 courses (37%). Grade III vomit-
`ing was observed in only 3 patients (3 of 99 courses
`(3%)) and included the patient with ascites and paralyt-
`ic ileus, and hence the toxicity may have been disease
`and not drug related.
`
`Renal toxicity
`
`In most pa..ents renal function (GFR) was not altered
`by lometrexol treatment. However, 2 of 8 patients treat-
`ed at 45 mg/m2 every 4 weeks, experienced a decrease
`in GFR of > 20% following a single course of treat-
`ment. One of these two patients, who received a second
`course, had a. further reduction from 65% of pretreat-
`ment following cycle I (164 to 107 ml/min), to 45%
`or pretreatment after cycle 2 (107 to 74 ml/rnin). In
`addition, one patient treated at 100 mg/rnz every 3
`weeks, had a reduction in GFR to 59% of pretreatment
`after cycle 2 (117 to 107 ml/min following cycle 1
`and to 69 ml/min following cycle 2). However, both
`of these last two patients were receiving concomitant
`oral non-steroidal anti-inflammatory drugs (diclofenac
`
`Previous preclinical experiments have demonstrated
`that leucovorin can reverse the cytotoxicity of lome-
`trexol [2, I3, 14]. Therefore leucovorin, initially at
`30 mg i.v., was instituted for patients who devel-
`oped 2 grade III toxicities other than alopecia or nau-
`sea. Severe toxicity which needed leucovorin rescue
`occurred in 1 patient following the first course of lome-
`trexol and in 2 patients following the second course.
`Thus leucovorin was given to a patient who developed
`grade III diarrhoea following the first course of lome-
`trexol at 45 mgimz every 4 weeks and the diarrhoea
`resolved within 48 hours of leucovorin administration.
`
`The two patients who developed grade IV thrombocy-
`topenia after their second course of lometrexol, one at
`30 mg/m2 and one at 45 mg/ml every 4 weeks (the sec-
`cud patient having previously received leucovorin after
`the first course due to diarrhoea), were treated with 30
`mg intravenous leucovorin every 6 hours for 12 days
`and 14 days, respectively, at which time myelosup-
`pression resolved (Figure 2). However, as there was no
`elem evidence that leucovorin had reversed lometrexol
`
`toxicity, patients who developed severe toxicity at dose
`levels > 45 mg/m2 were not treated with leucovorin.
`
`Antitumour effects
`
`An objective partial response was observed in one
`patient with metastatic breast cancer after her first
`course of lometrexol at 30 mg/m2. The response was
`in a soft tissue lesion and was sustained for 48 days.
`One patient treated at 45 mg/m2 every 3 weeks
`metastatic breast cancer achieved a minor response,
`with an improvement in unmeasurable skin metastases
`and reduced dyspnoea for 10 weeks (but no significant
`
`Lilly Ex. 2031
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2031
`Sandoz v. Lilly IPR2016-00318
`
`

`
`
`
`-—h-E‘.I—Z
`H-
`
`
`
`330
`
`Plateletsx1000percumm
`
`C)
`
`E;
`
`20
`
`30
`
`Days
`
`a A
`
`ran
`T
`
`on
`
`I
`l-I- Pt 10 (30mglsq.m) —g—_ Pt 16 (45mglsq.m)
`I:l
`l_..______.__.__..-_______l
`
`Figun 2. Lometrexol-induced thrombocytopenia in two patients treated with lencovorin. Patients 10 and 16 received 2 courses oflometrexol at
`30 and 45 mg/tn’ every 4 weeks. respectively. The timing of lometrexol administration is represented by arrows. Patient 10 received leucovotin
`alter the second course at the platelet count nadir, represented by the open bar (days 44-55). Patient 16 received Ieucovorin aflertlre first course
`(days 7-10), because of grade III diarrhoea, and alter the second course at the platelet count nadir (days 43-56), represented by the solid bars.
`
`change in lung lesions on chest x-ray). Four patients (1
`metastatic malignant melanome, 1 ovarian cancer and
`2 unknown primary carcinoma) had disease stabiliza-
`finrr Far 4 1 1 and < mnntlne no RGQPGEPI‘ H1: nhvcingl
`5:95.: .9.
`., .,, - ...... ., ....,......., ... ..............-s. .,_, r.._, ...-....
`
`examination and computed tomography scan.
`
`Plasmafolate statusfollowing oralfolic acid
`supplementation
`
`Plasma folate concentrations were measured in 23
`
`patients prior to folate supplementation (day —7) and
`after 7 days of 5 mglday oral folic acid, i.e. prior to
`lometrexol administration (day 0). Plasma folate con-
`centrations were significantly increased from 9 (3-63)
`to 20 (6—l80) ng/ml [median (range)] after 7 days of
`folate supplementation (paired t—test p = 0.009). Pre
`and post folate supplementation plasma folate con-
`centrations were variable, increasing in most but not
`all cases following the first week of folate supple-
`mentation by l0 (-6 to 134) nglml [median (range)],
`amounting to a 100% (-20% to 140095) change [medi-
`an (rangc)].
`
`Phannacolcincrics
`
`For the dose range 12-45 mg/m2, as previously
`descrikd {I2}, there was a linear relationship &!‘.'.'€‘.‘.!!
`lometrexol dose and area under the plasma concentra-
`tion versus time curve. In the present study, this rela-
`tionship was maintained at doses up to and including
`130 mg/m2 (r2 = 0.89).
`
`Dkcussion
`
`The objective of the present clinical study was to iden-
`tify a safe dose of lometrexol when given with folate
`supplementation so as to allow Phase II trials, in an
`attempt to reproduce the eflicacy of lometrexol seen
`in folate—deficient mice receiving folate supplemen-
`tation. The _most common and severe toxicities in this
`study were thrombocytopenia and mucositis. Leucope-
`nia, neuuopsaia, anaemia and diarrhoea were mild and
`infrequent. Reductions of GFR of 21-50% and > 50%
`after treatment with lometrexol were also observed in 3
`
`Lilly Ex. 2031
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2031
`Sandoz v. Lilly IPR2016-00318
`
`

`
`331
`
`LometrexolAUC(ug/ml.min)
`
`100
`
`I
`
`II
`
`010 20
`
`so
`
`40
`
`50
`
`so
`
`70
`
`so
`
`so 1oo11o12o13o14o
`
`Lometrexol Dose (mgIsq.m)
`
`Figure 3. The relationship between lometrexol dose and area under the lometrexol plasma concentration versus time curve (AUC). Each symbol
`is a separate patient and the line is that given by linear regression analysis.
`
`and 1 patients, respectively. However, no patient in this
`study developed renal toxicity as classified by WHO
`criteria.
`
`Thrombocytopenia was the most severe side-effect
`and was the dose limiting toxicity in patients with
`bowel dysfunction. The two patients who experienced
`grade IV thrombocytopenia after 2 courses of lome-
`trexol concurrently developed paralytic ileus and bowel
`obstruction due to primary disease, and this might have
`effected folate absorption. In these two patients plas-
`ma folate concentrations prior to and after one week
`of folate supplementation, but still before lometrexol
`administration, were 4 and 21 ng/ml (the patient at 30
`mg/m2 every 4 weeks) and 9 and 17 ng/ml (the patient
`at 45 mglmz every 4 weeks). it is important to note
`that plasma folate concentrations after supplementa-
`tion in these two patients were not very high, i.e. still
`only equal to or less than the upper limit of normal (21
`ng/ml). The plasma folate concentrations during peri-
`ods of thrombocytopenia and bowel dysfunction could
`not be determined because lometrexol interferes with
`
`the plasma folate assay that was used, i.e. one based
`on the folate binding protein. Studies in normal adults
`have shown reduced serum folate levels after only 3
`nu-snlrn run on
`I-'nIo-on Jnfininnr Aha» Tl E1 and lap,-an aha
`VV\.A4h.D U11 II l\Jlfll«U'|-|\vll\¢l\uIlL KLIUI. LIJJ, Illlll IIDIIUU L115
`
`two thrombocytopenic patients might have been folate
`deficient at the time toxicity developed due to their con-
`
`current gastrointestinal disease effecting folate absorp-
`tion. Therefore lometrexol with folate supplementation
`is not recommended for patients with abnormal bowel
`function or patients who have a tendency to devel-
`op obstruction. Grade III—IV thrombocytopenia was
`observed on 3 of9 courses (33%) at 130 mg/m2 which
`suggests that this toxicity may become dose limiting.
`Mucositis was observed in patients at 2 30 mglmz
`lometrexol every 4 weeks and was dose related. As
`one patient at 60 mglm’ and one patient at 170 mym’
`every 3 weeks developed grade IH mucositis. this side-
`effect may also become a dose limiting toxicity. Grade
`III diarrhoea was observed in only 4 courses; however,
`it is important to note that a number of patients had
`constipation before receiving lometrexol due to con-
`current narcotic analgesic treatment which may have
`masked subsequent diarrhoea.
`In addition to antiproliferative toxicities this study
`has demonstrated, for the first time, that nephrotoxicity
`is a side effect of lometrexol. For example, one patient
`at 45 mym’ every 4 weeks had a progressive reduc-
`tion in GFR to 45% of pretreatment after receiving two
`courses oflometrexol, even though the GFR at this time
`was still above the normal range (i.e. > 50 ml/min;
`reduction from 164 to 74 ml/min). However,
`it is
`important to note that from 7 days after the first course
`until 4 weeks after the second course of lometrexol
`
`Lilly Ex. 2031
`Sandoz V. Lilly IPR2016-00318
`
`Lilly Ex. 2031
`Sandoz v. Lilly IPR2016-00318
`
`

`
`332
`
`this patient also received diclofenac, a non steroidal
`anti-flammatory drug (NSAID), which can efi’ect renal
`function and cause acute renal failure [16]. Although
`it appears that NSAIDs have little effect on the GFR
`in normal controls [17], in high renin states includ-
`ing cirrhosis. NSAID treatment was found to decrease
`GFR [18]. Although the patient studied here did not
`have cirrhosis, the unknown primary adenocarcinoma
`with extensive liver metastasis may have effected liver
`function, and hence predisposed the patient to toxicity
`from diclofenac. A drug interaction between lome-
`trexol and NSAIDs has not been described previously.
`However, both high-dose [19] and intermediate-dose
`methotrexate [20] have been reported to cause acute
`renal failure and life-threatening toxicities when giv-
`en with NSAIl)s, which was associated with impaired
`methotrexate elimination [19]. A possible mechanism
`for the interaction between methotrexate and NSAIDs
`
`might be inhibition of renal prostaglandin synthesis
`by NSAIDs which would decrease renal perfusion and
`thus reduce methotrexate clearance. From the present
`study it is not possible to determine whether diclofenac
`itself, or lometrexol, or the combination of 2 drugs, was
`responsible for the renal dysfunction.
`Measurements of urinary lometrexol concentra-
`tions in patients at 12 to 45 mg/m2 every 4 weeks
`suggested dose related increases in the renal excretion
`of lometrexol [T2]. Lometrexol urinary excretion with-
`in the first 24 hours in patients treated with 45 mg/ml
`every 4 weeks accounted for over 90% of the dose
`administered [12], and the presence of large amounts
`of lometrexol in the urine may have been related to the
`onset of renai toxicity. When considering the mech-
`anism of renal toxicity, it is of interest to compare
`lometrexol with methotrexate and CB37 17, a classical
`
`antifolate inhibitor of thymidylate synthase. Both com-
`pounds have been shown to cause crystalluric tubular
`darnage as a result of their insolubility at the acidic pH
`(< pH 5.7; [22]) generally present in the renal tubule
`[22—25]. In the light of his information the solubility
`of lometrexol in urine was determined and this showed
`
`that at 1 mg/ml lometrexol was completely dissolved
`at pH 3 5. but that solubility was < 1 mg/ml at pH <
`5. If lometrexol renal toxicity were to become a clini-
`cally significant problem it may be possible to alkalin-
`ize patients using intravenous bicarbonate, a technique
`currently employed with high dose methotrexate ther-
`apy [26].
`Preclinical studies clearly demonstrated that the
`cytotoxic effects of lometrexol on tumour cells can
`be prevented by leucovorin [2, 13, 14]. In early Phase
`
`I clinical studies it was reported that the toxicities of
`lometrexol could be reversed within one week by leu-
`covorin (100 mglday), whereas they lasted for 7-49
`days in patients who did not receive leucovorin rescue
`[6]. In the present study, the severe thrombocytope-
`nia observed in two patients resolved after 2 weeks of
`leucovorin rescue. It is important to point out that the
`reversal of toxicity in this study took longer than report-
`ed by Ray and colleagues which may have been due to
`the higher doses of lometrexol used here. Alternative-
`ly, leucovorin may not in fact be effective in reversing
`the severe or late toxicities of lometrexol, or the dose
`of leucovorin used may not have been sufficient. When
`considering the dose of leucovorin used it is of interest
`to compare the amount given in this study (30 mg i.v.
`every 6 hours) to that used for high-dose methotrexate
`with leucovorin rescue. With high-dose methotrexate.
`leucovorin doses of 50-100 mg/m2 every 6 hours suc-
`cessfuly rescue patients with methotrexate levels above
`9 x 10*’ M at 48 hours, whereas patients with sim-
`ilarly high methotrexate levels who received 12-30
`mg/m2 of leucovorin still experienced severe myelo-
`suppression [27].
`in vim) studies have shown
`Mechanistically,
`that
`leucovorin,
`i.e. 5-formyl
`tetrahydrofolate,
`is
`metabolized intracellularly to the various one-
`carbon substituted folate cofactors
`including 5-
`methyltetrahydrofolate,
`l0-formyltetrahydrofolate,
`5,10-methylene tetrahydrofolate, tetrahydrofolate and
`their polyglutamate derivatives [28]. Interestingly, 10-
`fonnyltetrahydrofolate, which competes with lome-
`trexol for GARPT, was found to constitute only 15
`and 32% of totai intraceiiuiar foiates in human breast
`
`and colon cell lines, respectively, and a rapid efflux of
`a large portion of intracellular folates was observed 2
`hours after removal of 5-formyltetrahydrofolate. Sim-
`ilarly,
`in vivo in murine intestinal epithelial cells,
`leucovorin led to only a transient elevation of 10-
`formyltetrahydrofolate levels,
`in comparison to the
`increases seen in other reduced folate cofactor pools
`[29]. These results raise the possibility that the limit-
`ed efficacy of leucovorin in rescuing lometrexol tox-
`ici