`H u m a n A n t i - V a s c u l a r E n d o t h e l i a l G r o w t h F a c t o r i n
`P a t i e n t s W i t h A d v a n c e d C a n c e r
`
`By M.S. Gordon, K. Margolin, M. Talpaz, G. W. Sledge, Jr, E. Holmgren, R. Benjamin, S. Stalter,
`S. Shak, and D.C. Adelman
`
`Purpose: We investigated the safety and pharmaco-
`kinetics of a recombinant human monoclonal antibody
`to vascular endothelial growth factor (rhuMAb VEGF) in
`patients with cancer.
`PatientsandMethods: Cohorts of patients with met-
`astatic cancer having failed prior therapy entered a
`phase I trial of rhuMAb VEGF administered by a 90-
`minute intravenous infusion at doses from 0.1 to 10.0
`mg/kg on days 0, 28, 35, and 42. Patients underwent
`pharmacokinetic sampling on day 0 and had serum
`samples obtained during the subsequent 28 days. Re-
`sponse assessment was carried out on days 49 and 72.
`Results: Twenty-five patients with a median Eastern
`Cooperative Oncology Group performance status of 0
`were accrued. There were no grade III or IV adverse
`events definitely related to the antibody. There were
`three episodes of tumor-related bleeding. Infusions of
`
`rhuMAb VEGF were well tolerated without significant
`toxicity. Grades I and II adverse events possibly or
`probably related to study drug included asthenia,
`headache, and nausea. Pharmacokinetics revealed a
`linear profile with a half-life of 21 days. There were no
`objective responses, though 12 patients experienced
`stable disease over the duration of the study.
`Conclusion: rhuMAb VEGF was safely administered
`without dose-limiting toxicity at doses ranging up to 10
`mg/kg. Multiple doses of rhuMAb VEGF were well
`tolerated, and pharmacokinetic studies indicate that
`doses of > 0.3 mg/kg have a half-life similar to that of
`other humanized antibodies. Subsequent trials will ex-
`plore rhuMAb VEGF alone and in combination
`chemotherapy.
`J Clin Oncol 19:843-850. © 2001 by American
`SocietyofClinicalOncology.
`
`A NGIOGENESIS, OR new blood vessel formation, is
`
`critical to tumor growth, invasion, and metastasis.1
`Several humoral factors stimulate angiogenesis. These fac-
`tors act either by inducing the enzymatic breakdown of the
`perivascular basement membrane or by inducing prolifera-
`tion and chemotaxis of endothelial cells. Both components
`are critical for successful neovascularization, and the inhi-
`bition of either arm has been hypothesized as having a
`potential antitumor or antimetastatic effect on malignant
`cells. Vascular endothelial growth factor (VEGF) is a 43- to
`46-kd glycoprotein that induces the proliferation and migra-
`tion of vascular endothelial cells.2,3 These activities are
`mediated via the two receptors for VEGF, flt-1 and KDR,
`which are found predominantly on vascular endothelial
`cells.2 In preclinical models, VEGF is a potent neovascu-
`larization
`agent
`for
`both
`normal
`and malignant
`microvasculature.4,5
`Many malignant cells produce VEGF, which serves as an
`autocrine factor for the induction of neovascularization.
`Several studies have demonstrated a correlation between
`high levels of VEGF and increased risk of metastatic
`disease and overall poor prognosis in a variety of malignan-
`cies including non–small-cell lung cancer and other cancers.
`In addition, increased expression of VEGF by malignant
`tumors is associated with a more invasive phenotype.6-9 In
`preclinical animal models, the inhibition of VEGF is asso-
`ciated with stabilization of established tumors.10 When
`
`administered in conjunction with chemotherapy, a synergis-
`tic antitumor activity can be seen in preclinical models.11
`Recombinant human monoclonal antibody (rhuMAb)
`VEGF is a humanized monoclonal antibody that was gen-
`erated by engineering the VEGF binding residues of a
`murine neutralizing antibody into the framework of a
`normal human immunoglobulin G (IgG).12 This antibody
`binds and neutralizes all biologically active forms of VEGF
`(including VEGF165, VEGF121, and the thrombin split
`fragment VEGF110), because it recognizes the binding sites
`for the two VEGF receptors. The use of anti-VEGF anti-
`bodies has been extensively studied in preclinical in vivo
`models and has demonstrated an inhibition of tumor growth
`in a dose-dependent manner.13 We now report on the first
`phase I study with anti-VEGF, which was performed to
`evaluate its safety and pharmacokinetic profile in patients
`
`From the Indiana University School of Medicine, Indianapolis, IN;
`City of Hope National Medical Center, Duarte; Genentech, Inc, South
`San Francisco, CA; The University of Texas, M.D. Anderson Cancer
`Center, Houston, TX.
`Submitted October 28, 1999; accepted September 28, 2000.
`Funded by Genentech, Inc, and supported in part by Public Health
`Service grant no. MO1 RR750.
`Address reprint requests to Michael S. Gordon, MD, Suite 415, 4001
`N Third St, Phoenix, AZ 85012; email: msgordon@u.arizona.edu.
`© 2001 by American Society of Clinical Oncology.
`0732-183X/01/1903-843
`
`Journal of Clinical Oncology, Vol 19, No 3 (February 1), 2001: pp 843-850
`
`843
`
`Downloaded from ascopubs.org by Reprints Desk on November 30, 2016 from 216.185.156.028
`
`Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
`
`NOVARTIS EXHIBIT 2015
`Breckenridge v. Novartis, IPR 2017-01592
`Page 1 of 8
`
`
`
`844
`
`with relapsed and refractory malignancies. These studies
`demonstrate that rhuMAb VEGF is safe in the doses and
`schedule used here and that serum concentrations attained
`with both single and multiple doses successfully reproduce
`concentrations necessary for antitumor activity based on
`preclinical models.
`
`PATIENTS AND METHODS
`
`Inclusion Criteria
`
`through July 31, 1997, 25 patients with
`From May 1, 1997,
`measurable or assessable solid tumor malignancies were enrolled onto
`this phase I trial. Eligibility criteria included refractory advanced solid
`tumors for which no standard curative therapy existed, Eastern Coop-
`erative Oncology Group (ECOG) performance status # 1, normal
`hematologic function as demonstrated by an absolute neutrophil count
`greater than 1,500 cells/mL, hemoglobin greater than 9 g/dL (transfu-
`sion allowed), and a platelet count greater than 100,000/mL, as well as
`normal renal function (creatinine less than 1.5 mg/dL) and hepatic
`function (bilirubin , 1.5 times the upper limit of institutional normal).
`Patients were excluded if they had a known history of CNS metastatic
`disease with evidence of residual recurrent disease at study entry, had
`received chemotherapy or immunotherapy within the prior 4 weeks
`before study entry, or had taken any noncorticosteroidal anti-inflam-
`matory agents within 10 days of study entry. Patients were also
`excluded if they had undergone invasive surgical procedures including
`organ biopsies within 2 weeks of study entry or were pregnant or
`lactating. The institutional review boards for the three participating
`centers approved the protocol, and voluntary written informed consent
`was obtained from all patients.
`
`Study Drug Formulation and Administration
`
`RhuMAb VEGF was supplied as a clear to slightly opalescent, sterile
`liquid ready for parenteral administration. Each 100-mg (10 mg/mL)
`glass vial contained rhuMAb VEGF with histidine, trehalose, polysor-
`bate 20, and sterile water for injection, USP, pH 5.5. Vials contained no
`preservative and were for single use only. Appropriate concentrations
`of rhuMAb VEGF were diluted into D5W for infusion. Patients
`received their infusion of rhuMAb VEGF over 90 minutes by calcu-
`lated pump and underwent evaluation of vital signs including blood
`pressure, pulse, respiratory rate, and temperature before treatment, at
`intervals during infusion, and hourly for 3 hours after infusion. After
`their first infusion, patients were hospitalized for 24 hours, during
`which time they underwent serial pharmacokinetic sampling after
`infusion. During cycle 1 (days 0 to 28), patients underwent pharmaco-
`kinetic evaluation on day 0 as noted above and then subsequently had
`samples drawn on days 2, 4, 7, and 10 and weekly during routine visits
`for the duration of the study. After subsequent infusions on days 28, 35,
`and 42, patients were observed for 3 hours and subsequently discharged
`for outpatient follow-up. All patients were seen weekly during the 10
`weeks of study therapy and follow-up and underwent evaluation with
`physical examination including ECOG performance status, vital signs,
`and laboratory evaluation with complete blood count with manual
`differential, chemistry evaluation, prothrombin time/partial thrombo-
`plastin time, and urinalysis. Toxicities were monitored using the
`National Cancer Institute Common Toxicity Criteria adjusted for
`biologic response modifiers.
`Response assessment using either radiographic or physical exami-
`nation evaluation was carried out on days 49 and 72. Patients with
`
`GORDON ET AL
`
`objective responses were to be offered continued therapy on a separate
`extension study.
`
`VEGF and Anti-VEGF Levels
`
`Serum rhuMAb VEGF concentrations were determined using an
`enzyme-linked immunosorbent assay (ELISA), which uses truncated
`recombinant human VEGF for capture and a goat antibody to human
`IgG conjugated to horseradish peroxidase for detection. Concentrations
`of less than 78 ng/mL were considered less than reportable (LTR).
`Measurement of the serum levels of VEGF was performed with the
`ELISA using a monoclonal antibody to the heparin-binding domain of
`VEGF as both capture and detection. Therefore, it sees only full-length
`forms that contain this domain, ie, VEGF 165 and higher molecular
`weight forms. This format was chosen as it can detect free VEGF and
`VEGF bound to the therapeutic drug, rhuMAb VEGF. The LTR for this
`assay for VEGF is 20 pg/mL. Free VEGF was measured by passing
`serum through a Staphylococcus Protein A column to remove all IgGs,
`including antibody-bound VEGF. The flow-through fraction is mea-
`sured as free VEGF. Percentage free VEGF (% Free VEGF) is
`determined by using this free VEGF as a percentage of total VEGF as
`assayed in the unfractionated serum. Anti rhuMAb VEGF antibodies
`were assayed by ELISA using rhuMAb VEGF Fab for detection and a
`goat antibody to human IgG conjugated to horseradish peroxidase for
`detection; a titer of 2 was considered the sensitivity limit.
`
`Statistical Analysis
`
`This phase I study accrued five patients per dose level and
`planned to enroll an additional
`three patients if dose-limiting
`toxicity (defined as a grade III or greater adverse event using the
`biologic response modifier-adjusted common toxicity criteria) oc-
`curred in two patients in a given cohort. The toxic dose was defined
`as the dose level at which three or more patients in a given cohort
`experienced dose-limiting toxicity. The maximally tolerated dose
`was defined as one dose level below the toxic dose assuming that
`this level was well tolerated and fewer than two patients in the
`cohort experienced a dose-limiting toxicity.
`Comparison of VEGF, rhuMAb VEGF, and other laboratory studies
`were performed using a two-sided paired student’s t test. Individual and
`mean serum rhuMAb VEGF concentration-time data were plotted by
`dose group. Serum rhuMAb VEGF disposition was analyzed by
`compartmental methods. Individual parameter estimates were tabulated
`and summarized (mean, SD, range). RhuMAb VEGF pharmacokinetics
`was assessed for dose proportionality by graphic examination.
`Serum VEGF concentration-time data were analyzed by noncom-
`partmental methods and summarized by time and dose groups. Results
`are presented as the mean, SD, and minimum and maximum values.
`
`RESULTS
`
`Patient Characteristics
`
`Twenty-five patients (eight male, 17 female) were ac-
`crued to this study. All were eligible and assessable for
`safety. Only one patient, treated at the 3-mg/kg dose level,
`did not receive all four doses of rhuMAb VEGF because of
`a hemorrhage into a previously undiagnosed cerebral me-
`tastasis during the month after the single dose administra-
`tion. The diagnoses and demographic data are presented in
`Table 1. The median ECOG performance status was 0
`
`Downloaded from ascopubs.org by Reprints Desk on November 30, 2016 from 216.185.156.028
`
`Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
`
`NOVARTIS EXHIBIT 2015
`Breckenridge v. Novartis, IPR 2017-01592
`Page 2 of 8
`
`
`
`PHASE I TRIAL OF RHUMAB VEGF
`
`Table 1. Patient Characteristics
`
`Characteristic
`
`Total patients
`Men/women
`Age, years
`Median
`Range
`ECOG performance score
`0
`1
`Cancer type
`Sarcoma
`Renal
`Breast
`Lung
`Other
`Prior therapy
`Chemotherapy
`Radiation therapy
`Immunotherapy
`
`No. of Patients
`
`25
`8/17
`
`55
`21-70
`
`17
`8
`
`8
`7
`5
`2
`3
`
`22
`10
`10
`
`(range, 0 to 1), and the mean age was 51 years (range, 21 to
`70 years).
`
`Safety
`
`In general, rhuMAb VEGF was well tolerated at all doses
`studied. There were no Common Toxicity Criteria (CTC)
`grade 3 or 4 infusion-related toxicities. A small number of
`patients developed grade 1 or 2 adverse events characterized
`by asthenia, headache, nausea, or low-grade fever on the
`first day of rhuMAb VEGF administration (Table 2). Ad-
`verse events over the course of the entire study were similar
`in nature and predominantly of grades 1 to 2 in severity.
`These events are outlined in Table 3. Fever occurred in 10
`patients, though the relationship to the study drug adminis-
`tration could not be determined in all cases. There was no
`relationship between the severity of the fever and dose of
`the rhuMAb VEGF.
`No clinically significant changes were seen in biochem-
`ical, coagulation, or hematologic parameters. Although
`surgical interventions were limited to necessary procedures
`only, no patient demonstrated objective impairment of
`wound healing as a result of rhuMAbVEGF therapy. Minor
`changes in blood pressure were noted to be associated with
`rhuMAb VEGF administration. Systolic and diastolic blood
`pressures in patients treated at the 3 and 10 mg/kg dose
`levels increased an average of more than 10 mm Hg at some
`point during therapy. No significant changes in other vital
`signs were noted.
`Adverse events graded as 3 or 4 on the CTC scale
`occurred in four patients (Table 3). These included a patient
`with anemia at the 0.1 mg/kg dose level and one patient with
`
`Table 2. Adverse Events Occurring in Over 20% of Patients on rhuMAb
`VEGF (all grades and attributions for 25 patients treated)
`
`845
`
`Adverse Event
`
`Asthenia
`Headache
`Fever
`Rash
`Oral symptoms
`Nausea
`Arthralgias
`Pain
`Cough
`Emesis
`Dyspnea
`
`No. of Subjects
`
`Grade 1-2
`
`Grade 3-4
`
`18
`11
`10
`9
`8
`7
`7
`7
`6
`6
`5
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`1
`
`dyspnea at the 0.3 mg/kg dose level. In both of these cases,
`the events were attributable to progression of the patient’s
`underlying malignancy. In addition, there were two epi-
`sodes of serious bleeding, both at the 3.0 mg/kg dose level.
`The first of these patients was a 29 year-old female with a
`history of hepatocellular carcinoma. The patient had under-
`gone a previous trisegmentectomy and subsequently devel-
`oped multiple pulmonary metastases. She was treated with
`combination chemotherapy including carboplatin, doxoru-
`bicin, and cyclosporine with her best response being pro-
`gressive disease. She received her first dose of rhuMAb
`VEGF at a dose of 3.0 mg/kg and on day 14 of cycle 1 was
`bicycling when she experienced a grand mal seizure and an
`acute cerebrovascular accident. She was evaluated with a
`CT scan of the head that demonstrated a cerebrovascular
`bleed and underwent emergent surgery for the evacuation of
`the hemorrhage. Pathologic evaluation of the surgical spec-
`imen revealed residual hepatocellular carcinoma consistent
`with hemorrhage into a previously unrecognized brain
`metastasis. An extensive review of the literature revealed a
`high-rate of tumor associated hemorrhage as the presenting
`sign in up to 87.5% in one series.14 Based on these findings,
`it was decided in conjunction with the sponsor that the event
`was disease-related.
`The second patient was a 38-year-old female with a
`primary diagnosis of an epithelioid sarcoma of the right
`thigh. Sites of disease included a large right thigh mass and
`multiple pulmonary metastases. She had received extensive
`prior therapy with multiple chemotherapy regimens as well
`as external beam radiation therapy and brachytherapy. On
`approximately study day 39, she noted increasing pain and
`swelling in her right thigh with discoloration of the tumor
`area. This area continued to expand and eventually ruptured
`resulting in a severe hemorrhagic complication requiring
`local therapy for control. This patient also experienced an
`
`Downloaded from ascopubs.org by Reprints Desk on November 30, 2016 from 216.185.156.028
`
`Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
`
`NOVARTIS EXHIBIT 2015
`Breckenridge v. Novartis, IPR 2017-01592
`Page 3 of 8
`
`
`
`846
`
`Dose (mg/kg)
`
`Grade
`
`Anemia
`Anxiety
`Constipation
`Diarrhea
`Dysphagia
`Dyspnea
`Edema
`HTN
`IC bleed
`Myalgias
`Nightmares
`Sinus
`Sweats
`Tumor Hem
`
`0.1
`
`1-2
`
`3-4
`
`1
`
`2
`1
`
`1
`1
`
`Table 3. Adverse Event Profile (all grades)
`
`0.3
`
`1.0
`
`3.0
`
`10.0
`
`1-2
`
`3-4
`
`1-2
`
`3-4
`
`1-2
`
`3-4
`
`1-2
`
`3-4
`
`GORDON ET AL
`
`1
`
`1
`
`1
`
`1
`
`2
`
`1
`
`2
`
`1
`2
`
`1
`
`1
`1
`
`1
`
`2
`
`1
`
`2
`
`1
`
`1
`
`NOTE. All toxicities are listed, regardless of their potential relationship to study drug administration.
`Abbreviations: HTN, hypertension; IC, intracerebral; Hem, hemorrhage.
`
`uncomplicated episode of hemoptysis on day 57. Both of
`these episodes were related to the necrosis of existing
`tumors and were not believed to be reflect adverse events
`related to the study drug.
`Based on the dose escalation schema defined in the
`protocol, expansion of the cohort was deemed as indicated
`if the two serious adverse events occurred in the first 28
`days of the study. Although the event related to the CNS
`metastasis bleed was within this 28-day period, the second
`occurred beyond this point and therefore did not qualify to
`indicate a need for cohort expansion.
`Two other patients (liposarcoma and breast cancer) re-
`ported episodes of minor hemoptysis. These occurred on
`days 57 and 2 of therapy, respectively, and spontaneously
`resolved. Both patients had recognized pulmonary metasta-
`ses, and in both cases, it was believed that the bleeding was
`related to their underlying disease, though an association to
`the study drug could not be ruled out. Neither of the two
`premenopausal women experienced menstrual abnormali-
`ties during or after participation in this study.
`
`Efficacy
`
`No patient treated on this phase I study experienced an
`objective partial or complete response. One patient with
`renal cell carcinoma, treated at the 10 mg/kg dose level,
`experienced a minor response with an approximately 20%
`to 30% reduction in the sum of perpendicular diameters of
`pulmonary and lymph node metastases. Among 23 patients
`who were assessable for response at 70 days, 12 experi-
`enced stable disease over the 70-day study interval, with the
`remaining 11 patients demonstrating progressive disease.
`The patients with stable disease included five with renal cell
`
`cancer and were otherwise distributed among the other
`previously noted diagnoses. Aside from the higher number
`of patients with renal cell cancer, no other definable
`association between stabilization and sites of metastases,
`age, sex, or prior therapy could be identified. The small
`numbers of patients, heterogeneity of tumor types, patient
`characteristics and durations of therapy, and lack of an
`established definition of stable disease preclude the deter-
`mination of a meaningful association between the dose of
`rhuMAb VEGF dose and disease stability. Baseline serum
`VEGF levels in the patients with stable disease ranged from
`LTR to 281 pg/mL, with a mean of 98.4 compared with
`those patients with progressive disease who had baseline
`values of LTR to 122 pg/mL with a mean of 41.6. The
`patient with the minor response treated at the 10.0 mg/kg
`dose level was followed off therapy and progressed within
`4 to 5 months of completion of treatment. He was subse-
`quently retreated with rhuMAb VEGF and demonstrated
`another minor response with shrinkage of multiple pulmo-
`nary metastases and mediastinal nodal disease that lasted for
`8 months until new bone metastases were identified. One
`additional patient with renal cell cancer developed objective
`minor regression of multiple hepatic metastases after
`completion of rhuMAb VEGF therapy (no change in an
`intact renal primary tumor). This response lasted 11
`months before progressive disease with new bone metas-
`tases were identified.
`
`Antibodies to rhuMAb VEGF
`
`No patient enrolled onto the trial developed antibodies to
`rhuMAb VEGF during the period of measurements (70 days).
`
`Downloaded from ascopubs.org by Reprints Desk on November 30, 2016 from 216.185.156.028
`
`Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
`
`NOVARTIS EXHIBIT 2015
`Breckenridge v. Novartis, IPR 2017-01592
`Page 4 of 8
`
`
`
`PHASE I TRIAL OF RHUMAB VEGF
`
`Table 4. Pharmacokinetic Profile for Single Dose rhuMAb VEGF
`
`Cmax (mcg/ml)
`
`CL (ml/kg/day)
`
`MRT
`
`AUCinf
`(day*mg/mL)
`
`0.1 mg/kg
`1
`2
`3
`4
`5
`0.3 mg/kg
`6
`7
`8
`9
`10
`1.0 mg/kg
`11
`12
`13
`14
`15
`3.0 mg/kg
`16
`17
`18
`19
`20
`10.0 mg/kg
`21
`22
`23
`24
`25
`
`3.97
`1.92
`2.68
`1.44
`2.37
`
`5.61
`6.18
`7.72
`4.8
`8.91
`
`29.9
`39.3
`23
`24.6
`21
`
`ND
`89.6
`75
`52.5
`91.5
`
`186
`206
`294
`277
`294
`
`2.3
`16.4
`3.8
`16.6
`6.55
`
`5.94
`4.7
`3.83
`9.58
`3.45
`
`4.17
`2.62
`3.87
`4.13
`2.95
`
`ND
`1.75
`2.07
`5.46
`4.5
`
`4.02
`4.06
`1.91
`2.44
`1.66
`
`10.9
`3.14
`9.79
`4.16
`7.77
`
`8.98
`10.3
`10.9
`6.49
`15.7
`
`7.99
`16.7
`17.1
`17.6
`24.6
`
`ND
`19.1
`30
`10.4
`12.8
`
`13.3
`12
`17.7
`27.2
`54.3
`
`43.5
`6.09
`26.3
`6.03
`15.3
`
`50.5
`63.8
`78.3
`31.3
`86.9
`
`240
`382
`259
`242
`339
`
`ND
`1720
`1450
`550
`666
`
`2480
`2490
`5230
`4100
`6010
`
`Abbreviations: Cmax, maximal concentration; CL, clearance; MRT, mean
`resonance time; AUCinf, area under the curve; ND, not done.
`
`Pharmacokinetic Studies of rhuMAb VEGF
`
`After administration of the first rhuMAb VEGF dose,
`mean observed Cmax ranged from 2.80 mg/mL for the 0.1
`mg/kg group to 284 mg/mL for the 10 mg/kg group (Table
`4). These changes were dose-related and there was no
`significant accumulation of rhuMAbVEGF during the mul-
`tidosing portion of the study (data not shown). Mean kinetic
`profiles of the rhuMAb VEGF pharmacokinetics for the
`multiple administration portion of the study are show in Fig 1.
`The mean rhuMAb VEGF clearance for the 0.1 mg/kg
`dose group (9.29 mL/kg/d) was higher than the clearance for
`all other dose groups (range 2.75-5.07 mL/kg/d); the larger
`mean resulted primarily from two of the patients whose
`clearances were greater than 14 mL/kg/d. Clearance values
`for the other three subjects were consistent with those
`estimated at higher doses. Over the range of doses of 0.3 to
`10.0 mg/kg, the kinetics of rhuMAb VEGF seems to be
`linear, with a t1/2 of approximately 21 days. Overall, the
`pharmacokinetic profile indicates that when rhuMAb VEGF
`
`847
`
`Fig 1. Mean serum rhuMAb VEGF concentrations. Serum levels of
`rhuMAb VEGF after serial administrations (days 0, 28, 35, and 42) at doses
`of 0.1 (closed triangle), 0.3 (open square), 1.0 (closed square), 3.0 (open
`triangle), and 10.0 (closed circle) mg/kg. Cohorts consist of 4 to 5 patients.
`
`was administered once followed by a 28-day washout
`period and then weekly for 3 weeks at doses ranging from
`0.1 to 10 mg/kg, the disposition was characterized by a low
`clearance and a volume of distribution consistent with
`limited extravascular distribution.
`
`Serum Levels of VEGF
`Before rhuMAb VEGF administration, individual serum
`VEGF concentrations ranged from less than 20 to 281
`pg/mL. The two patients with major hemorrhagic events had
`pretreatment serum VEGF levels of 30.6 and 122 pg/mL,
`respectively. The latter of these was slightly elevated
`compared with the mean values across the different dose
`levels, though higher baseline levels were seen in a number
`of patients. Among the seven patients with renal cell
`carcinoma, the baseline serum VEGF concentrations ranged
`from LTR to 218 pg/mL (median 56.9 pg/mL). An increase
`in serum total VEGF concentration was observed across all
`dose groups; the increase was more consistent with doses of
`greater than 1.0 mg/kg with serum levels two to four times
`higher for the 1.0, 3.0, and 10 mg/kg dose groups than for
`the 0.1 and 0.3 mg/kg dose groups (Table 5). Free serum
`VEGF concentrations were found to be reduced and, at
`doses of $ 0.3 mg/kg, were below the detectable limit of the
`assay after the administration of rhuMAb VEGF and re-
`mained undetectable for the duration of the study (data
`based on eight patients not shown; personal communication,
`2000, D. Fei, PhD, Genentech, Inc, South San Francisco, CA).
`
`DISCUSSION
`The use of antiangiogenic agents as anticancer therapy
`has been the focus of numerous clinical investigations over
`the past several years. The ability to inhibit neovasculariza-
`tion and prevent
`tumor growth and metastases has the
`
`Downloaded from ascopubs.org by Reprints Desk on November 30, 2016 from 216.185.156.028
`
`Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
`
`NOVARTIS EXHIBIT 2015
`Breckenridge v. Novartis, IPR 2017-01592
`Page 5 of 8
`
`
`
`848
`
`0.1 mg/kg
`1
`2
`3
`4
`5
`Mean
`0.3 mg/kg
`6
`7
`8
`9
`10
`Mean
`1.0 mg/kg
`11
`12
`13
`14
`15
`Mean
`3.0 mg/kg
`16
`17
`18
`19
`20
`Mean
`10.0 mg/kg
`21
`22
`23
`24
`25
`Mean
`
`Table 5. Serum VEGF Concentrations (pg/mL) and Relationship to Tumor Diagnosis and Response to Therapy
`
`GORDON ET AL
`
`Serum VEGF
`Baseline
`
`40
`56.9
`25.2
`23.5
`27.1
`34.5 6 14.1
`
`176
`LTR
`56.6
`76.3
`LTR
`69.8 6 64
`
`28.8
`126
`68.7
`119
`87.7
`86 1 39.6
`
`30.6
`27.5
`41.4
`122
`39.1
`52.1 6 39.5
`
`281
`120
`LTR
`218
`39.4
`136 6 107
`
`Serum VEGF
`Day 72
`
`ND
`67.2
`LTR
`37.7
`84.1
`52.3 6 23.5
`
`235
`41.7
`29.8
`70.6
`LTR
`79.4 6 95.4
`
`203
`159
`126
`256
`96.6
`168 1 63.1
`
`ND
`230
`ND
`171
`135
`179 6 48
`
`140
`412
`685
`497
`488
`444 6 198
`
`Tumor Type
`
`NSCLC
`Renal
`Breast
`Renal
`Sarcoma
`
`Sarcoma
`Renal
`Sarcoma
`Sarcoma
`Sarcoma
`
`Breast
`NSCLC
`Breast
`Renal
`Renal
`
`Hepatoma
`Sarcoma
`Renal
`Sarcoma
`Breast
`
`Sarcoma
`H&N
`Breast
`Renal
`Melanoma
`
`Best
`Response at
`Day 70
`
`Stable
`Progression
`Progression
`Progression
`Progression
`
`Stable
`Stable
`Progression
`Progression
`Progression
`
`Progression
`Stable
`Stable
`Stable
`Stable
`
`SAE
`Stable
`Stable
`Progression
`Stable
`
`Stable
`Stable
`Progression
`Stable
`Stable
`
`Abbreviations: LTR, below detectable limits; NSCLC, non–small-cell lung cancer; H&N, head and neck; SAE, severe adverse event.
`
`potential to open a new forum for the treatment of cancer.
`Vascular endothelial growth factor (VEGF) is a potent
`inducer of endothelial cell proliferation and migration.5
`Increased serum levels of VEGF have been associated with
`poor prognosis in a variety of malignancies, which further
`enhances the hypothesis that VEGF may serve as an
`autocrine growth factor
`for malignant neovasculariza-
`tion.15,16 In addition, the deregulation of VEGF as is seen
`with many cases of sporadic renal cell carcinoma or in von
`Hippel-Lindau syndrome related to mutations in the VHL
`gene indicate a possible pathophysiologic role for VEGF in
`certain malignancies.17,18
`We performed a phase I trial of a recombinant human
`monoclonal antibody directed against VEGF. Pre-clinical
`studies with this molecule have demonstrated an excellent
`
`safety profile with the inhibition of ovulation and slight
`growth retardation in female and male cynomolgus mon-
`keys, respectively, as the only evident adverse events. We
`saw a similarly excellent safety profile with no dose-
`limiting toxicity experienced by our patients at doses
`ranging from 0.1 to 10 mg/kg. Adverse events including
`asthenia, headache, and nausea were seen in a minority of
`patients and were mild in nature. Mild increases in systolic
`and diastolic blood pressures (10 to 15 mm Hg) were seen
`at the 3 and 10 mg/kg dose levels. The specific mechanism
`of action for this effect remains unclear and further research
`will need to be conducted to elucidate this mechanism if
`subsequent trials demonstrate this as a reproducible effect.
`Several bleeding episodes, all tumor-related were seen in
`this study. It is unclear what role the inhibition of VEGF
`
`Downloaded from ascopubs.org by Reprints Desk on November 30, 2016 from 216.185.156.028
`
`Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
`
`NOVARTIS EXHIBIT 2015
`Breckenridge v. Novartis, IPR 2017-01592
`Page 6 of 8
`
`
`
`PHASE I TRIAL OF RHUMAB VEGF
`
`may have played in these cases, though an effect on rapidly
`proliferating tumor vascularity cannot be completely ruled
`out. Minor bleeding problems, such as hemoptysis in
`patients with pulmonary metastases, was seen in two pa-
`tients but was self-limiting in both cases. There have been
`no preclinical models demonstrating an increased risk of
`bleeding or thrombosis associated with VEGF inhibition.
`Recently presented clinical trials, using various inhibitors of
`VEGF, have suggested a potential increased risk of bleeding
`or thrombosis in specific clinical settings.19-21 The exact
`mechanism and its potential relationship to VEGF inhibition
`remain to be defined. Kubo et al22 have recently published
`data demonstrating that VEGF is involved in the mainte-
`nance of endothelial cell integrity of tumor microvascula-
`ture. The blockade of VEGF receptor-3 results in an
`increased rate of apoptosis of these endothelial cells leading
`to the exposure of subendothelial tissue that may trigger a
`coagulation cascade. It is certainly possible that tumor-
`related bleeding may in part be related to such a phenom-
`enon though further studies will need to be carried out to
`define an association.
`There was no clear association between elevated baseline
`serum VEGF levels and risk of bleeding though the small
`number of cases precludes any statistical analysis of this
`association. The potential for a synergistic effect when
`rhuMAbVEGF is combined with effective chemotherapy
`where rapid shrinkage of tumors may occur needs to be
`considered with regard to risk of bleeding.
`After the administration of a single dose of rhuMAb
`VEGF, pharmacokinetic studies demonstrated a half-life of
`approximately 21 days at doses $ 0.3 mg/kg. This pharma-
`cokinetic profile is similar to that of other humanized
`monoclonal antibodies using a similar Fc backbone. The
`pharmacokinetics following multiple weekly dosing dem-
`onstrated a slight accumulation of rhuMAb VEGF concen-
`trations though the kinetics remained linear. In conjunction
`with these studies, serum VEGF concentrations were also
`evaluated. Although increases in serum total VEGF concen-
`trations were seen, this is likely a result of an increase in
`VEGF synthesis/distribution and/or a decrease in VEGF
`clearance caused by complex formation between VEGF and
`rhuMAb VEGF. This latter effect is also seen with other
`antibodies such as the rhuMAb-E25 that binds and inacti-
`vates IgE. The half-life of IgE is shorter than that of the
`complex, and consequently, total circulating IgE rises four-
`to five-fold in patients on treatment.23 Serum free VEGF
`levels are dramatically reduced with the first dose of
`rhuMAb VEGF at doses $ 3.0 mg/kg and may indicate a
`potentially important biologic property of this therapy if the
`same pharmacodynamic effect
`is
`seen in the tumor
`microenvironment.
`
`849
`
`Although antitumor response was not a primary objective
`of the study, we evaluated all patients for potential antitu-
`mor activity of this therapy. Although no objective re-
`sponses were seen, two patients had a minor response,
`suggesting potential antitumor activity. In addition, 12 of 23
`patients experienced stable disease during the 70-day period
`of the study. It is interesting that five of these patients had
`renal cell cancer, a disease that in its sporadic form is
`characterized by elevated expression of VEGF related to the
`deregulation of VEGF degradation resulting from mutations
`in the VHL gene. The renal cell patients on this study had
`baseline serum VEGF levels similar to those of the other
`diagnoses accrued. It was interesting to note, however, that
`patients with stable disease seemed to have mildly higher
`baseline serum VEGF levels compared with those patients
`with progressive disease. Unfortunately, this study did not
`treat enough patients at any single dose to adequately
`characterize a relationship between response and baseline
`VEGF profiles. It is interesting to hypothesize, however,
`that elevated endogenous VEGF concentrations may indi-
`cate tumors that are more VEGF driven and, therefore, may
`be better targets for VEGF inhibition. Evaluation of this
`relationship will need to be explored in single agent or
`combination phase II trials. It is certainly possible that more
`prolonged or higher-dose exposure to antiangiogenic agents
`will be necessary to induce objective responses or mean-
`ingful prolongation of progression-free survival in patients
`with established tumors, and hence, we cannot rule out the
`possibility that further treatment of stable patients could
`result in objective tumor shri