`c 1997 Kluwer Academic Publishers. Printed in the Netherlands.
`
`235
`
`A phase I trial and pharmacokinetic evaluation of CI-980 in patients with
`advanced solid tumors
`
`Nancy T. Sklarin1, Chetan D. Lathia2, Laura Benson1, William R. Grove2, Sylvia Thomas1,
`Javier Roca1, Avi I. Einzig1 and Peter H. Wiernik1
`1Albert Einstein Cancer Center, Bronx, New York; and 2Parke-Davis Pharmaceutical Research, Division of
`Warner-Lambert Company, Ann Arbor, Michigan, USA
`
`Key words: CI-980, Phase I, solid tumors
`
`Abstract
`
`CI-980 is a synthetic mitotic inhibitor that binds to the colchicine binding site of tubulin. It demonstrates broad
`activity against human and murine tumor models and shows no cross resistance with tumor models whose mechanism
`of resistance is mediated by P-glycoprotein (MDR-1). A phase I study was completed in 25 patients with solid
`tumors using a 24-hour infusion schedule, with courses repeated every 3 weeks. Eight dose levels were tested
`between 1.2 and 15.6 mg/m2. The maximum tolerated dose was 14.4 mg/m2. Neutropenia was dose-related but
`not dose-limiting; thrombocytopenia was infrequent. CNS toxicities were dose-limiting and consisted of dizziness,
`headache, loss of coordination, loss of consciousness, nervousness, and other symptoms. These events occurred
`near the end of the infusion and were reversible, usually within 24 hours. One patient who was to be treated at dose
`level 8 (intended dose was 19.2 mg/m2; actual dose was 15.6 mg/m2) became encephalopathic prior to completion
`of the infusion. Other adverse events included gastrointestinal toxicities (nausea, vomiting, anorexia, constipation,
`stomatitis, dyspepsia, bleeding, cheilitis), IV site erythema, fever, and fatigue. A partial response was observed in
`one patient with colon cancer and reductions in CA-125 levels were observed in 2 patients with ovarian cancer.
`Pharmacokinetics were linear and dose-proportional. Results indicate high systemic clearance and wide tissue
`distribution. Mean pharmacokinetic parameter values: T1=2 = 5.52 hours, plasma clearance 1163 mL/min/m2, and
`Vdss 376 L/m2.
`
`Introduction
`
`CI-980, ethyl (S)-(5-amino-1, 2-dihydro-2-methyl-
`3-phenylpyridol[3,4-b]pyrazin-7-yl)carbamate 2-hy-
`droxyethanesulfonate (1:1) (Figure 1), is a synthetic
`antineoplastic agent that acts as a mitotic inhibitor by
`blocking tubulin polymerization [1]. It binds to the
`colchicine binding site on tubulin, distinct from that
`of the vinca alkaloids and taxanes. CI-980 was more
`potent and more active than vincristine in in vitro test-
`ing against murine and human tumor cell lines. In vivo
`testing demonstrated similar activity when compared to
`vincristine in many tumors, including P388 and L1210
`leukemias, B16 melanoma, and several colon cancer
`lines [2].
`In preclinical studies, CI-980 retained activity
`regardless of the route of drug administration and the
`site of tumor implantation, a characteristic not found in
`
`Figure 1. CI-980 structure.
`
`currently approved mitotic inhibitors. Unlike the vin-
`ca alkaloids, no cross-resistance to CI-980 was found
`in a large number of drug-resistance tumor models,
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 1 of 12
`
`
`
`236
`
`whether tested in vitro or in vivo. This implies that CI-
`980 may not be susceptible to pleiotrophic drug resis-
`tance and therefore could potentially be a useful agent
`in patients previously treated with anthracyclines, vin-
`ca alkaloids, etc. CI-980 exhibited marked schedule
`dependency in vitro and in vivo. Using a clonogenic
`assay, long exposure times (i.e., 24 hours and con-
`tinuous) were approximately 105 times more effective
`than shorter exposure times (i.e., 1 and 4 hours) in
`inhibiting colony formation of L1210 cells. In addi-
`tion, various CI-980 treatment schedules were studied
`in mice implanted with P388 leukemia or mamma-
`ry 16/C tumor. The optimal schedules appeared to be
`those that provided a prolonged exposure to CI-980
`[3].
`In animals, the toxicologic profile of CI-980 was
`consistent with the action of mitotic inhibitors, affect-
`ing primarily rapidly dividing cell populations. Target
`organ effects included bone marrow suppression, lym-
`phoid depletion, enteropathy, epithelial necrosis, tes-
`ticular tabular degeneration, and degenerative urinary
`bladder lesions [4]. In addition, endosteal new bone
`formation and stromal proliferation were observed in
`rats [5]. Peripheral neuropathy could not be evaluated
`in the animal studies but was an anticipated toxicity
`in the clinical trials, because of the similarity in the
`mechanism of action with vinca alkaloids.
`Based on these findings, four CI-980 Phase I studies
`have been conducted, each designed to provide a pro-
`longed period of exposure. One study treated patients
`over a 3-day period using intravenous infusions of var-
`ious durations [6, 7]. Two studies treated patients using
`a continuous intravenous infusion for 72 hours [8, 9].
`The study reported here utilized a 24-hour infusion
`of CI-980, with courses repeated every 3 weeks. The
`objectives of this study were to determine the toler-
`ance and pharmacokinetic parameters of CI-980 when
`administered using this schedule in patients with solid
`tumors. Preliminary results have been presented [10].
`
`Patients and methods
`
`Patient selection
`
`Patients were eligible if they had a histologically con-
`firmed advanced solid tumor which was refractory to
`standard chemotherapy. Measurable or evaluable dis-
`ease was required, as was a performance status of 0–2
`(ECOG) and an expected survival of at least 8 weeks.
`Laboratory values were obtained within one week pri-
`
`or to the start of CI-980 therapy and were required
`to be as follows: BUN 30 mg/dL, serum creatinine
` 1.5 mg/dL, bilirubin and SGOT < 1.5 times the
`upper limit of normal, normal serum calcium, sodi-
`um 130 mEq/L, white blood count 3.0 103/l,
`absolute granulocyte count 1.5 103/l, platelet
`count 100 103/l. All patients were informed of
`the investigational nature of the study and provided
`written informed consent. The study was approved by
`the institutional review board of Montefiore Medical
`Center.
`Due to the similarity in the mechanism of action
`between CI-980 and the vinca alkaloids and taxanes,
`it was anticipated that CI-980 might cause peripher-
`al neuropathy. Therefore, to allow an assessment of
`CI-980’s toxicities,
`limitations were placed on the
`amount of prior chemotherapy allowed with these
`agents. Patients were excluded if they had received pri-
`or treatment with more than one vinca alkaloid, a vinca
`alkaloid and paclitaxel, or had a clinically significant
`peripheral neuropathy. Also excluded were patients
`with brain metastases unless they were clinically stable
`for 4 weeks. Because acute hypertension was reported
`in another Phase I study [7], patients were excluded if
`they had hypertension that required therapy with agents
`other than thiazides. Patients were also excluded if they
`had received chemotherapy within 4 weeks prior to the
`start of the study, or mitomycin or nitrosourea thera-
`py within 6 weeks. Small port radiotherapy within 4
`weeks was allowed for symptomatic relief.
`
`Patient evaluation
`
`Preliminary patient evaluation included a history and
`physical examination including a detailed neurolog-
`ic review. Baseline studies included a complete blood
`count, electrolytes, renal and liver function tests, cal-
`cium magnesium, chest x-ray, ECG, and appropriate
`scans to identify disease sites. Levels of tumor markers
`were analyzed where applicable. Interim blood counts
`were obtained three times between treatment and day
`15 of each course, and chemistry and electrolyte values
`were checked weekly. A chest x-ray was obtained on
`day 1 of each course in most patients. Relevant x-rays
`and scans were repeated after every 2–3 cycles to fol-
`low the course of the disease. During the first course,
`patients underwent blood pressure monitoring using an
`ambulatory blood pressure monitor.
`A complete response was defined as complete dis-
`appearance of all known disease for at least 28 days,
`a partial response as 50% decrease in the sum of
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 2 of 12
`
`
`
`the products of the greatest perpendicular diameters
`of bidimensionally measurable lesions, and progres-
`sive disease as a 25% increase in any tumor lesion.
`Toxicities were graded according to NCI Common-
`Toxicity Criteria. The MTD was defined as that dose
`which produced dose-limiting toxicity in at least one-
`half of the patients initially treated at that dose level.
`Nonhematologic dose-limiting toxicity was defined as
`grade 3 or 4 toxicity, or grade 1 or 2 toxicity lasting
` 6 weeks. Hematologic dose-limiting toxicity was
`defined as a granulocyte count nadir 0.5 103/l, a
`platelet count nadir 50 103/l, or failure to recover
`to 1.5 103/l granulocytes or to 100 103/l
`platelets within 6 weeks after treatment. Depending on
`the severity of toxicities encountered, the recommend-
`ed Phase II dose would be either 75% or 100% of the
`MTD.
`
`Drug formulation and administration
`
`CI-980 was supplied by Parke-Davis Pharmaceutical
`Research, Division of Warner-Lambert Company (Ann
`Arbor, MI), as a sterile preservative-free lyophilized
`powder containing 10 mg of CI-980 base equivalent.
`The drug was reconstituted with Water for Injection to
`a concentration of 2 mg/mL and then further diluted
`with 5% Dextrose in Water to a concentration of 50 to
`500 g/mL, depending on the dose level. Doses were
`prepared in a 60cc Becton-Dickinson polyethylene
`syringe, primed through a Baxter microvolume tubing
`extension set, and administered via a Medfusion 2001
`syringe pump over 24 hours. Because CI-980 adheres
`to certain plastics, infusions were administered through
`a peripheral vein using Teflon angiocaths.
`The starting dose for this Phase I study, 1.2 mg/m2,
`was determined from a series of toxicology studies.
`This dose was equivalent to 1/10 the LD10 following
`administration of a single intravenous bolus dose in
`mice, and was also the no-effect dose following admin-
`istration of either a single intravenous bolus dose or a
`24-hour continuous intravenous infusion in dogs (0.06
`mg/kg).
`Three new patients were treated at each dose level.
`Doses were escalated based on the modified Fibonac-
`ci escalation scheme. Courses were repeated every
`3 weeks provided the patient recovered from drug-
`attributable adverse events. In the absence of progres-
`sive disease or toxicity, patients received the highest
`currently open dose level in subsequent courses.
`
`237
`
`Sample collection
`
`CI-980 pharmacokinetics were studied during the first
`treatment course. Heparinized blood samples were col-
`lected at 0, 4, 8, 16, 23.5, 24 (end of infusion), 24.5,
`25, 26, 28, 32, 36 and 48 hours following the start of
`the infusion. Plasma were harvested after centrifuga-
`tion and the resulting samples were frozen at -70 C
`until analyzed. Urine samples were collected in Teflon
`or stainless steel containers at intervals of 0–6, 6–12,
`12–24, 24–36, and 36–48 hours following the start
`infusion. In a few patients, urine was collected in two
`intervals: 0–24 and 24–48 hours. Total volume of urine
`collected over each interval was recorded and a 5 mL
`aliquot was removed, stored in Teflon vials, and frozen
`at -70 C until analyzed.
`
`Analytical methods
`
`Plasma CI-980 concentrations were determined with
`a validated HPLC assay with fluorescence detection
`[11]. Following the addition of PD 080658 (inter-
`nal standard-IS) samples were vortexed and applied
`to Bond-ElutTM C-18 cartridges preconditioned with
`methanol and water. Cartridges loaded with samples
`were washed with acetonitrile mixed with 0.2% ammo-
`nium acetate buffer (30:70). CI-980 and IS were elut-
`ed using a 60:40 mixture of acetonitrile and 0.2%
`ammonium acetate buffer. The eluate was evaporated
`to dryness and reconstituted in mobile phase. CI-980
`was resolved from the IS using two Zorbax Rx C-18
`columns connected in series and maintained at 30 C
`and a mobile phase of acetonitrile and 10 mM ammo-
`nium dihydrogen phosphate buffer (38:62). Detection
`was achieved using a fluorescence detector with an
`excitation wavelength of 388 nm and an emission
`wavelength of 473 nm. The range of the calibration
`curve was 0.2–25 ng/mL. Accuracy and precision of
`quality controls during validation of the assay were
`within 4.7% and 5.6%, respectively.
`Urine CI-980 concentrations were determined in
`a manner similar to those of plasma concentrations.
`The main difference was that 1 mL of serum albu-
`min (BSA) solution was added to every 5 mL of urine
`before the addition of PD 080658. Following this the
`IS was added and the samples were vortexed for 30 sec-
`onds. Urine samples were processed in the same fash-
`ion as plasma. The eluate from solid phase extraction
`was evaporated to dryness and reconstituted in mobile
`phase which is 40:60 of acetonitrile mixed with 10
`mM ammonium dihydrogen phosphate buffer. A Zor-
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 3 of 12
`
`
`
`238
`
`bax Rx C-18 column maintained at 30–35 C was used
`to chromatograph the compound and internal standard.
`Detection was achieved using a UV detector at a wave-
`length of 380 nm. The range of calibration curve was
`1–100 ng/mL. Accuracy and precision of quality con-
`trols during validation of assay were within 1.5% and
`4.3%, respectively.
`
`Pharmacokinetic methods
`
`CI-980 pharmacokinetic parameters were determined
`from CI-980 plasma and urinary concentration-time
`data using noncompartmental methods. Maximum
`plasma CI-980 concentration (Cmax) was recorded as
`observed. AUC(0–tldc) and AUMC(0–tldc), the AUC
`and AUMC from time zero to the time of the last
`detectable plasma concentration (ldc), respectively,
`were determined by Lagrange polynomial interpola-
`tion using LAGRAN [12]. The apparent elimination
`rate constant (z) was estimated as the absolute val-
`ue of the slope of the least squares linear regression
`line of the natural logarithm (ln) of plasma concentra-
`tions as a function of time during the terminal phase of
`drug disposition. Apparent elimination half-life (T1=2)
`values were calculated as 0.693/z. AUC(0– ) val-
`ues were determined by summing AUC(0–tldc) and
`ldc/z values. AUMC(0– ) values were determined
`from the sum of AUMC(0–tldc) and (ldc/z2 + ldc (cid:0)
`tldc/z). Total plasma clearance (CLTOT) was calcu-
`lated as dose/AUC(0- ). Mean residence time (MRT)
`was calculated as ((AUMC(0– )/AUC(0– ))-(T/2))
`where T is the length of the constant intravenous infu-
`sion. Volume of distribution at steady state was cal-
`culated as CLTOT-MRT. Renal clearance (CLR) was
`determined by dividing the amount of CI-980 excreted
`in urine from 0 to 24 hours (Ae(0–24)) by the AUC
`from 0 to 24 hours (AUC(0–24)). CI-980 blood clear-
`ance was calculated as the product of the calculated
`plasma clearance and the ratio of plasma to blood con-
`centrations.
`
`Pharmacodynamic analysis
`
`The pharmacodynamics of CI-980 were evaluated to
`correlate CNS toxicity and myelosuppression with
`some measure of CI-980 plasma exposure. CNS
`adverse event data were analyzed as a function of Cmax,
`dose, and the time the plasma CI-980 concentration
`was above 7 ng/mL. The relationships between phar-
`macokinetic parameters (dose, AUC(0 (cid:0) ), Cmax,
`CLTOT) and the grade of CNS toxicity or precent
`
`Table 1. Patient characteristics
`
`Male/female
`Median age (range)
`Performance status
`0
`1
`2
`
`Primary tumor type
`Adenocarcinoma of colon
`Renal cell
`Sarcoma
`Ovarian
`Small cell lung
`Nonsmall cell lung
`Squamous cell esophagus
`Squamous cell head/neck
`Squamous cell vulva
`Melanoma
`Mesothelioma
`Breast
`
`Prior treatment
`None
`Chemotherapy
`Chemotherapy + radiotherapy
`Radiotherapy
`
`15/10
`62 (30–78)
`
`2
`19
`4
`
`10
`3
`2
`2
`1
`1
`1
`1
`1
`1
`1
`1
`
`1
`12
`11
`1
`
`decrease in neutrophils from baseline values were eval-
`uated by linear regression analysis.
`
`Results
`
`Treatment
`
`Twenty-five patients with 11 different solid tumor diag-
`noses participated in this trial. Patient characteristics
`are shown in Table 1. This was a heavily pretreated
`patient population, having received a median of three
`chemotherapy regimens prior to entering the study. All
`patients met the eligibility requirements described in
`“Patients and methods”.
`One hundred-twelve courses of CI-980 treatment
`were administered over the dose range 1.2 to 15.6
`mg/m2 (range 1 to 24 courses per patient, median 3).
`Using the modified Fibonacci dose escalation scheme,
`8 dose levels were required to reach the MTD (Table 2).
`Three new patients were treated at each dose level until
`toxicities at dose level 8 became apparent. Additional
`patients were then treated at dose level 6 (1 patient)
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 4 of 12
`
`
`
`Table 2. CI-980 treatment dose and schedule
`
`Dose Dose
`(mg/m2)
`level
`
`Number of
`Total
`New
`Percent
`increase# patients patients courses
`
`1.2
`2.4
`4.0
`6.0
`8.4
`10.8
`14.4
`19.2
`
`1
`2
`3
`4
`5
`6
`7
`8
`Total
`
`– –
`100
`67
`50
`40
`29
`33
`33
`
`3
`4
`4
`6
`9
`13
`9
`1
`
`3
`3
`3
`3
`3
`4
`5
`1
`25
`
`5
`8
`6
`10
`17
`26
`39
`1
`112
`
` Actual total dose administered = 15.6 mg/m2.
`# Percent increase over prior dose level.
`
`and dose level 7 (2 patients) to confirm the toxicity
`findings.
`Twelve patients received escalated doses and two
`patients received deescalated doses in subsequent
`courses. Six patients remained on study for an extend-
`ed treatment of 7 courses, including one patient who
`received 24 courses of CI-980 (total dose of 568 mg).
`
`Adverse events
`
`Neurotoxicity. Acute reversible neurotoxicity princi-
`pally affecting the central nervous system (CNS) was
`the dose-limiting toxicity. The most common neuro-
`logic adverse events were those suggestive of cerebel-
`lar dysfunction, including dizziness, loss of coordina-
`tion, ataxia, and tremors. Eleven of 25 patients (44%)
`reported neurologic adverse events. Many of these
`events were first reported after the decision was made
`to deescalate the dose following the adverse events
`described below at the 19.2 mg/m2 dose level.
`Table 3 shows the dose-relationship of neurologic
`adverse events that were considered to be related to
`study treatment. The lowest dose that was associat-
`ed with any neurologic adverse event was 6.0 mg/m2,
`where three separate grade 1 events were reported. At
`10.8 mg/m2, several grade 1 and 2 neurologic adverse
`events were reported, including dizziness, headache,
`ataxia, loss of coordination, and one grade 3 event (loss
`of consciousness). The largest dosing experience was
`obtained at the 14.4 mg/m2 dose level where a total of
`39 courses of treatment were given. Seven of 9 patients
`treated at this dose level reported at least one neuro-
`logic adverse event, including three grade 3 events of
`dizziness, loss of consciousness, and hand tremors.
`
`239
`
`One patient was treated at the 19.2 mg/m2 dose level
`and experienced multiple neurologic adverse events,
`described below.
`CNS symptoms usually began during or within one
`day following the CI-980 infusion. All CNS adverse
`events were reversible, usually within a few days after
`completion of the infusion. Neurotoxicity commonly
`occurred after the first course of treatment and did not
`appear to be cumulative in those patients who were
`treated with multiple courses.
`Two patients experienced a total of 3 episodes of
`loss of consciousness. One patient’s initial episode
`occurred on day 2 following treatment at 14.4 mg/m2,
`and a second episode occurred on day 2 follow-
`ing a subsequent course of treatment at 10.8 mg/m2.
`The patient quickly recovered consciousness follow-
`ing both episodes. There was no seizure activity or
`incontinence associated with these events. Another
`patient began CI-980 treatment at the 19.2 mg/m2
`dose level and experienced significant CNS toxici-
`ties. Within several hours of starting the infusion, the
`patient described a band-like pain across the chest and
`abdomen, followed by severe peripheral vasoconstric-
`tion, diaphoresis, agitation, confusion, and hyperten-
`sion. Because of worsening symptoms, the CI-980
`infusion was terminated after 19 hours, delivering an
`estimated total dose of 15.6 mg/m2. Within one-half
`hour of discontinuing the infusion, the diaphoresis and
`skin changes resolved. However, during the next day
`the patient’s neurologic condition deteriorated,exhibit-
`ing hyperreflexia, flacidity, a right gaze preference, and
`Babinski signs. The patient’s level of consciousness
`fluctuated over the ensuing four days before recovering
`completely. Because there was a reduction in this ovar-
`ian cancer patient’s CA-125 levels following this first
`CI-980 treatment, the patient insisted on being retreat-
`ed. She experienced no recurrence of CNS adverse
`events following additional treatment courses at 10.8
`mg/m2.
`
`dose-
`Myelosuppression. Meylosuppression was
`related but not dose-limiting (Table 4). Neutropenia
`did not occur following doses of 1.2–8.4 mg/m2. At
`14.4 mg/m2, grade 3 or 4 neutropenia occurred in 4 of
`39 courses. Nadirs occurred early, usually between day
`6 and 12 of each course; the median time to recovery
`was day 14. Thrombocytopenia was infrequent and
`generally was not clinically significant. One patient
`treated at 14.4 mg/m2 and one patient treated at 15.6
`mg/m2 developed grade 2 thrombocytopenia.
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 5 of 12
`
`
`
`240
`
`Table 3. Treatment-related neurotoxicitiesa , by dose-level
`
`Adverse
`event
`
`6.0 mg/m2
`(10
`courses)
`
`8.4 mg/m2
`(17
`courses)
`
`10.8 mg/m2
`(26
`courses)
`
`14.4 mg/m2
`(39
`courses)
`
`19.2 mg/m2
`(1
`courseb)
`
`1
`
`1
`1
`
`Dizziness
`Headache
`Paresthesia
`Incoordination
`Unconscousness
`Neuropathy
`Hypesthesia
`Hypertonia
`Nervousness
`Ataxia
`Emotional lability
`Strange dreams
`Speech disorder
`Confusion
`Agitation
`Tremor
`
`[1]
`
`1
`2
`1
`1
`1
`
`1
`1
`
`2
`
`1
`1
`
`1
`
`7
`5
`1
`3
`1
`3
`
`1
`
`1
`1
`1
`
`1
`
`[1]
`
`[1]
`
`[1]
`
`1
`1
`1
`
`1
`
`[1]
`
`[1]
`
`a Numbers represent the number of times an adverse event was reported. Numbers in brackets
`represent the number of times an adverse event of Grade 3 severity was reported.
`b Actual dose administered was 15.6 mg/m2
`
`Table 4. Neutropenia by dose level
`
`Dose level
`mg/m2
`
`Number of
`patients
`
`Number of maximum WHO
`course
`Toxicity Grade
`
`1.2
`2.4
`4.0
`6.0
`8.4
`10.8
`14.4
`19.2
`
`3
`4
`4
`6
`9
`13
`9
`1
`
`5
`8
`6
`10
`17
`26
`39
`1
`
` Actual dose administered 15.6 mg/m2.
`
`0
`
`5
`8
`6
`10
`17
`25
`22
`
`1
`
`2
`
`3
`
`4
`
`1
`7
`
`6
`
`1
`
`3
`1
`
`Other adverse events. Table 5 summarizes the inci-
`dence and severity of reported nonhematologic adverse
`events other than neurologic events. Most adverse
`events were grade 1 or 2 in severity. The only events
`reported with grade 3 severity were nausea/vomiting
`and dyspepsia. No grade 4 events occurred. Four
`episodes of gastrointestinal bleeding occurred without
`coexisting thrombocytopenia. Three of these involved
`separate episodes of hematemesis and gastritis follow-
`ing courses 1 (14.4 mg/m2), 5 (14.4 mg/m2), and 7
`
`(10.8 mg/m2) in one patient with a history of bleeding
`peptic ulcers. At all dose levels, CI-980 infusions com-
`monly resulted in localized erythema or tenderness at
`the site of infusion and along the infused vein. Nausea,
`vomiting, fever, and fatigue were common following
`doses 10.8 mg/m2. Fever generally occurred during
`or within one day of receiving treatment. No treatment-
`related deaths occurred.
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 6 of 12
`
`
`
`Table 5. Treatment-related adverse events, by dose-level
`
`241
`
`Adverse
`event
`
`Nausea
`and/or
`vomittig
`IV site
`reaction
`Fever
`Fatigue
`Anorexia
`Constipation
`Stomatities
`Dyspepsia
`GI bleed
`Cheilitis
`
`1.2 mg/m2
`(5
`courses)
`
`2.4 mg/m2
`(8
`courses)
`
`4.0 mg/m2
`(6
`courses)
`
`6.0 mg/m2
`(10
`courses)
`
`8.4 mg/m2
`(17
`courses)
`
`10.8 mg/m2
`(26
`courses)
`
`14.4 mg/m2
`(39
`courses)
`
`19.2 mg/m2
`(1
`courseb)
`
`3
`
`3
`
`1
`
`2
`
`1
`
`3
`
`1
`
`1
`2
`
`5
`
`4
`
`4
`8
`4
`1
`1
`1
`1
`
`21
`
`[2]
`
`3
`
`20
`9
`3
`3
`6
`3
`2
`4
`
`[2]
`
`1
`
`1
`
`1
`
`a Numbers represent the number of times an adverse event was reported. Numbers in parentheses represent the number of times an
`adverse event of Grade 3 severity was reported.
`b Actual dose administered was 15.6 mg/m2.
`
`Pharmacokinetics
`
`Plasma CI-980 concentrations declined monoexponen-
`tially in most patients following the end of infusion. A
`plot of a typical plasma concentration-time profile at
`14.4 mg/m2 is presented in Figure 2. Mean (%RSD)
`pharmacokinetic parameters of CI-980 are presented in
`Table 6 by dose level. Individidual Cmax values ranged
`from 1.06 to 17.7 ng/mL. Individual AUC(0 (cid:0) )
`values ranged from 12.5 to 530 ng-hr/mL. Cmax and
`AUC(0 (cid:0) ) increased in proportion to dose over the
`dosage range of 1.2 to 15.6 mg/m2, consistent with lin-
`ear pharmacokinetics. A plot of AUC(0 (cid:0) ) vs. dose
`is presented in Figure 3. CLTOT did not change with
`increasing dose, which is also consistent with linear
`pharmacokinetics. Mean half-life, Vss, and CLTOT val-
`ues were 5.52 hours, 376 L/m2, and 1163 mL/min/m2,
`respectively. The volume of distribution at steady state
`of 376 L/m2 is greater than that of total body water
`suggesting extensive distribution into tissues. CI-980
`blood clearance calculated from the total plasma clear-
`ance value and the RBC binding was 1325 mL/min
`[13]. This value approaches human hepatic blood flow
`[14]. These findings are consistent with the pharma-
`cokinetics of CI-980 in the mouse and dog, which also
`showed rapid clearance of CI-980. Mean CLR value in
`these patients was 4.54 mL/min. A mean of 0.45% of
`the CI-980 dose was excreted unchanged in urine over
`the first 48 hours, suggesting that CI-980 is primari-
`
`ly cleared by nonrenal elimination. (The urine CI-980
`assay quantitated unchanged CI-980 only.)
`
`Pharmacodynamics
`
`CNS toxicity was the dose-limiting toxicity for CI-980
`with this dosing schedule. Pharmacodynamic relation-
`ships between grade of neurotoxicity and Cmax, dose,
`or duration above various (7 to 17 ng/mL) plasma CI-
`980 concentrations were explored by linear regression
`analysis. Of these variables, Cmax appeared to correlate
`best with the grade of neurotoxicity (r2 = 0.486) (Fig-
`ure 4). However, this correlation was not considered
`predictive enough to recommend prospective pharma-
`cokinetic monitoring in future clinical trials. Neither
`the dose (r2 = 0.331) nor the duration of exposure
`above 7 ng/mL (r2 = 0.230) appeared to correlate as
`well as Cmax. Duration of exposure above plasma CI-
`980 concentrations of 8 ng/mL (to 18 ng/mL) did
`not correlate well with CNS toxicity.
`The relationships between pharmacokinetic para-
`meters (dose, AUC, or Cmax) and the percent decrease
`in neutrophil counts with CI-980 treatment were eval-
`uated by linear regression analysis. Of these parame-
`ters, dose was the best predictor of percent decrease in
`neutrophil counts (r2 = 0.556), as shown in Figure 5.
`AUC(0 (cid:0) ) (r2 = 0.527) and Cmax (r2 = 0.447) also
`correlated well.
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 7 of 12
`
`
`
`242
`
`Figure 2. Plot of plasma CI-980 concentration-time data in a cancer patient dosed 14.4 mg/m2 CI-980 as a 24-hour intravenous infusion. The
`solid line represents a point-to-point connection of the observed data.
`
`Figure 3. Plot of plasma CI-980 AUC value as a function of dose following a 24-hour intravenous infusion of 1.2 to 15.6 mg/m2. Solid line
`represents the fit of a linear regression through the observed data. Linear regression R-squared value = 0.632.
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 8 of 12
`
`
`
`243
`
`Figure 4. CNS toxicity plotted as a function of plasma CI-980 Cmax value. Solid line represents the fit of a linear regression through the observed
`data. Linear regression R-squared value = 0.486.
`
`Figure 5. Percent decrease in absolute neutrophil counts versus dose in a phase I trial of CI-980. Solid line represents the fit of a linear regression
`through the observed data. Linear regression R-square value = 0.556.
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 9 of 12
`
`
`
`244
`
`Table 6. Mean (%RSD) pharmacokinetic parameters of CI-980 following a 24-hr intravenous infusion
`
`Dose
`(mg/m2)
`
`Cmax
`(ng/mL)
`
`AUC(0 (cid:0) )
`(ng*hr/mL)
`
`CLTOT
`(mL/min)
`
`CLTOT
`(ml/min/m2)
`
`CLR
`(mL/min)
`
`Vss
`(L)
`
`Vss
`(L/m2)
`
`% Urinary
`Excretion (0-48h)
`
`1.2
`f3g
`
`2.4
`f3g
`
`4.0
`
`6.0
`f3g
`
`1.27
`(21.6)
`
`1.32
`(19.3)
`
`2.12
`(27.9)
`f3g
`
`6.22
`(61.3)
`
`25.7
`(54.9)
`
`32.0
`(8.97)
`
`31.1
`
`(c)
`f1g
`
`103.0
`(39.5)
`
`1757.0
`(51.6)
`
`2444.0
`(10.7)
`
`3267.0
`(c)
`f1g
`
`1983.0
`(35.9)
`
`993.0
`(57.4)
`
`1253.0
`(9.58)
`
`2135.0
`(c)
`f1g
`
`109.3
`(44.7)
`
`1.20
`(15.3)
`
`5.01
`(67.9)
`
`3.89
`(85.9)
`f3g
`
`6.96
`(15.1)
`
`472.0
`(38.3)
`
`973.0
`(36.0)
`
`1207.0
`(c)
`f1g
`
`266.0
`(43.6)
`
`522.0
`(46.4)
`
`789.0
`(c)
`f1g
`
`0.338
`(28.9)
`
`0.323
`(29.8)
`
`0.314
`(140.0)
`f3g
`
`432.0
`(65.3)
`
`238.0
`(69.7)
`
`0.581
`(47.7)
`
`ta
`1=2
`hour
`
`4.76
`(65.1)
`
`9.55
`(12.6)
`
`7.20
`(c)
`f1g
`
`5.29
`(44.0)
`
`4.08
`(36.0)
`
`8.4
`f3g
`
`10.8
`f4g
`
`14.4
`
`15.6
`f1g
`
`8.48
`(57.8)
`
`5.45
`(38.2)
`
`11.4
`(29.7)
`f5g
`
`17.7
`(c)
`
`172.0
`(56.2)
`
`147.0
`(51.4)
`
`288.0
`(47.9)
`f5g
`
`399.0
`
`(c)
`
`Mean PK Parametersb
`
`1896.0
`(77.2)
`
`2527.0
`(38.5)
`
`1624.0
`(37.7)
`f5g
`
`1351.0
`(c)
`
`2047.0
`(41.7)
`f23g
`
`1103.0
`(72.9)
`
`1434.0
`(40.7)
`
`938.0
`(33.3)
`f5g
`
`804.0
`(c)
`
`1163.0
`(43.7)
`f23g
`
`5.83
`(74.6)
`
`3.41
`(84.2)
`
`5.98
`(122.0)
`f4g
`
`1.34
`(c)
`
`4.54
`(90.7)
`f24g
`
`330.0
`(78.8)
`
`846.0
`(50.4)
`
`647.0
`(38.8)
`f5g
`
`693.0
`(c)
`
`658.0
`(53.5)
`f23g
`
`191.0
`(74.9)
`
`484.0
`(52.1)
`
`374.0
`(35.0)
`f5g
`
`412.0
`(c)
`
`376.0
`(56.0)
`f23g
`
`0.627
`(75.0)
`
`0.350
`(59.4)
`
`0.660
`(72.7)
`f4g
`
`0.0815
`(c)
`
`0.453
`(74.8)
`f24g
`
`3.55
`(26.4)
`
`8.21
`(21.8)
`f5g
`
`21.3
`(c)
`
`5.52
`(73.2)
`f23g
`
`( ) = %RSD = relative standard deviation; f g = # of patients; a expressed as the harmonic mean; b values are derived from PK parameters of
`all individuals in the study; c not calculated.
`
`Responses
`
`Repeated evaluations of tumor lesions were obtained in
`patients with measurable disease. One partial response
`was documented in a patient with colon carcinoma with
`liver metastases. The response occurred after the third
`treatment course and lasted for eleven weeks. The CI-
`980 dose administered to this patient was reduced after
`the fifth course because of gastrointestinal toxicity. His
`CEA (carcinoembryonic antigen) level then began to
`rise and a CT scan revealed progressive disease.
`Two patients with ovarian carcinoma who had pre-
`viously failed paclitaxel were treated with CI-980. One
`patient had a baseline CA-125 level of 530 units/mL
`(normal value < 35 units/mL) which decreased to 136
`units/mL following treatment. Another patient had a
`CA-125 level of > 5000 units/mL obtained on day 8 of
`course 1. The CA-125 value decreased to 640 units/mL
`following the first course of treatment at 19.2 mg/m2,
`
`but then gradually increased during three subsequent
`courses of CI-980 treatment given at a reduced dose.
`
`Discussion
`
`This study evaluated the pharmacokinetics and toler-
`ance of CI-980 when administered as a 24-hour con-
`tinuous intravenous infusion, with courses repeated
`every 3 weeks. Based on preclinical toxicology find-
`ings, myelosuppression was expected to be the dose-
`limiting toxicity. However, CI-980 produced an unusu-
`al spectrum of adverse events that was not predicted by
`the toxicology findings. Because neurologic findings
`were the dose-limiting toxicity observed in this study,
`this subject will be the focus of the discussion.
`Neurotoxicity is commonly encountered in patients
`treated with mitotic inhibitors such as vincristine, vin-
`blastine and the taxanes [15, 16]. These drugs typically
`
`NOVARTIS EXHIBIT 2083
`Par v. Novartis, IPR 2016-01479
`Page 10 of 12
`
`
`
`produce peripheral neuropathy manifested as paresthe-
`sias and loss of reflexes due to axonal degeneration
`and demyelination. The onset of these effects is usu-
`ally delayed and result from cumualtive exposure fol-
`lowing several weeks or months of treatment. Central
`nervous system effects of vinca alkaloids, including
`seizures, disorientation, ataxia, and coma have been
`rarely reported [17]. Although some peripheral neu-
`ropathy was noted in this study, the more common
`and troublesome neurotoxicities involved the central
`nervous system.
`Eleven of 25 patients (44%) treated with CI-980
`developed some manifestation of neurotoxicity. In
`most patients, these adverse were central in nature
`rather than peripheral, and included complaints of
`dizziness, headache, loss of coordination, loss of con-
`sciousness, nervousness, and other symptoms. The
`occurrence of these adverse events was acute and clear-
`ly dose-related; both the frequency and the severity of
`CNS events increased with dose. These symptoms usu-
`ally appeared during treatment or within one day fol-
`lowing treatment and resolved within a few days. They
`did not appear to be the result of cumulative exposure
`to the drug.
`The potential for the development of CNS toxicities
`in man is consistent with data that demonstates CI-980
`can penetrate the blood-brain-barrier in animals (per-
`sonal communication). Autoradiography