Journal of Neuro-Oncology 62: 123–134, 2003.
`© 2003 Kluwer Academic Publishers. Printed in the Netherlands.
`
`Boron neutron capture therapy of brain tumors: clinical trials at
`the Finnish facility using boronophenylalanine
`
`Heikki Joensuu1, Leena Kankaanranta1, Tiina Sepp¨al¨a2, Iiro Auterinen3, Merja Kallio4, Martti Kulvik4,
`Juha Laakso5, Jyrki V¨ah¨atalo6, Mika Kortesniemi7, Petri Kotiluoto3, Tom Ser´en3, Johanna Karila2, Antti Brander8,
`Eija J¨arviluoma9, P¨aivi Ryyn¨anen10, Anders Paetau11, Inkeri Ruokonen5, Heikki Minn12, Mikko Tenhunen13,
`Juha J¨a¨askel¨ainen14, Markus F¨arkkil¨a4 and Sauli Savolainen15
`1Department of Oncology, 2Department of Physical Sciences, University of Helsinki; VTT Processes, VTT;
`3VTT Processes, VTT; 4Department of Neurology, and Clinical Research Institute; 5Department of Clinical
`Pharmacology; 6Laboratory of Radiochemistry, University of Helsinki; 7Department of Laboratory Diagnostics
`and Radiology, Helsinki University Central Hospital, Helsinki and Clinical Research Institute and Department of
`Physical Sciences, VTT Processes, VTT; 8Department of Radiology; 9Department of Pharmacy; 10Department of
`Physical Sciences, University of Helsinki; 11Department of Pathology; 12Turku PET Centre, Turku; 13Department
`of Oncology; 14Department of Neurosurgery; 15Department of Laboratory Diagnostics and Radiology, Helsinki
`University Central Hospital, Helsinki, Finland
`
`Key words: boron neutron capture therapy, BNCT, glioblastoma, glioma, neutron beam radiotherapy,
`radiotherapy
`
`Summary
`
`Two clinical trials are currently running at the Finnish dedicated boron neutron capture therapy (BNCT) facil-
`ity. Between May 1999 and December 2001, 18 patients with supratentorial glioblastoma were treated with
`boronophenylalanine (BPA)-based BNCT within a context of a prospective clinical trial (protocol P-01). All patients
`underwent prior surgery, but none had received conventional radiotherapy or cancer chemotherapy before BNCT.
`BPA-fructose was given as 2-h infusion at BPA-dosages ranging from 290 to 400 mg/kg prior to neutron beam irra-
`diation, which was given as a single fraction from two fields. The average planning target volume dose ranged from
`30 to 61 Gy (W), and the average normal brain dose from 3 to 6 Gy (W). The treatment was generally well tolerated,
`and none of the patients have died during the first months following BNCT. The estimated 1-year overall survival is
`61%. In another trial (protocol P-03), three patients with recurring or progressing glioblastoma following surgery
`and conventional cranial radiotherapy to 50–60 Gy, were treated with BPA-based BNCT using the BPA dosage of
`290 mg/kg. The average planning target dose in these patients was 25–29 Gy (W), and the average whole brain dose
`2–3 Gy (W). All three patients tolerated brain reirradiation with BNCT, and none died during the first three months
`following BNCT. We conclude that BPA-based BNCT has been relatively well tolerated both in previously irradi-
`ated and unirradiated glioblastoma patients. Efficacy comparisons with conventional photon radiation are difficult
`due to patient selection and confounding factors such as other treatments given, but the results support continuation
`of clinical research on BPA-based BNCT.
`
`Introduction
`
`The first patient was treated with boron neutron cap-
`ture therapy (BNCT) at the Finnish Research Reactor
`(FiR 1) in May 1999. The reactor is located within the
`Helsinki metropolitan area (about one million inhabi-
`tants) at Otaniemi, Espoo, about 6 km from the largest
`
`hospital of Finland, the Helsinki University Central
`Hospital. The FiR 1 reactor, a light-water moder-
`ated 250 kW Triga Mark II nuclear research reactor,
`was taken in use in 1962. It functioned as a training
`and research reactor for neutron activation analy-
`sis, isotope production, and neutron physics until the
`mid-1990s. In 1996, an epithermal neutron beam was
`
`CFAD v. Anacor, IPR2015-01776, CFAD EXHIBIT 1073 - Page 1 of 12
`
`

`
`124
`
`Figure 1. Schematic drawing of the BNCT facility at Fir 1.
`
`constructed based on a new neutron moderator mate-
`rial Fluental™ developed at VTT (Technical Research
`Centre of Finland) [1,2]. After successful demon-
`stration of a high purity epithermal beam, the patient
`irradiation room was constructed by cutting partly into
`the concrete shielding of the reactor (Figure 1). The
`Fluental™ moderator was shortened to create at that
`time the highest intensity and best purity epithermal
`neutron beam for BNCT. The whole reactor building
`was renovated including construction of irradiation
`simulation and monitoring rooms, and a laboratory for
`boron analysis, creating a dedicated clinical BNCT
`facility at the reactor site [3].
`Patients are treated in a collaboration with the
`Helsinki University Central Hospital, VTT, and the
`NC-Treatment Ltd. The Finnish BNCT multispecial-
`ity team consists of radiation therapists and clinical
`oncologists, neurologists, neurosurgeons, radiologists,
`pathologists, radiation physicists, chemists, pharma-
`cists, nurses, and the nuclear reactor facility personnel.
`The BNCT facility has been licensed for clinical
`
`use and is being surveyed by Finnish Nuclear and
`Radiation Safety Authority (STUK). The FiR 1 neu-
`tron beam is particularly well suited for BNCT because
`of its low hydrogen-recoil and incident gamma doses,
`and its high intensity and penetrating neutron spectrum
`characteristics [4].
`To improve patient safety and to further characterize
`the properties of the FiR 1 neutron beam, beagle dogs
`were irradiated with the FiR 1 beam before starting
`the current clinical trials. The beagles were irradia-
`ted using escalating neutron doses without the 10B
`carrier
`compound L-boronophenylalanine-fructose
`(L-BPA-F), but one dose group was infused with
`700 mg L-BPA/kg body weight. In these experiments
`the relative biological efficiency (RBE) of the FiR 1
`beam as compared with a conventional Linac 6 MV
`photon beam turned out to be about 1.25 in the dog
`brain [5]. Glioblastoma multiforme was chosen as the
`first tumor type to be treated, because treatment results
`achieved with conventional therapies are uniformly
`poor in this disease. Moreover, the pioneering work on
`
`CFAD v. Anacor, IPR2015-01776, CFAD EXHIBIT 1073 - Page 2 of 12
`
`

`
`125
`
`total
`B-10
`gamma
`N-14
`fast
`
`20
`
`18
`
`16
`
`14
`
`12
`
`10
`
`02468
`
`Weighted doses (Gy (W)
`
`0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
`Depth (cm)
`
`Figure 2. Modeled weighted dose rates to the normal brain in the
`central axis of a circular 14 cm diameter beam in a patient head.
`Boron concentration is 13 ppm.
`
`and gamma fields both in free beams and in phan-
`toms. Based on this model, a source description for
`the treatment planning code has been created.
`
`Boronophenylalanine-fructose
`
`Prior to clinical studies, different synthetic batches
`of L-BPA were investigated [11]. For clinical stud-
`ies L-BPA was purchased from Katchem Ltd. (Prague,
`Czech Republic). L-BPA was complexed under asep-
`tic conditions with fructose to form L-BPA-F at the
`pharmacy of the Helsinki University Central Hospital.
`The L-BPA-F solution was prepared at a concentra-
`tion of 30 g L-BPA/L by combining L-BPA with 10%
`molar excess of fructose in water. The final pH of the
`L-BPA-F solution was adjusted to 7.6 and tested for
`pyrogens before use. The infused amount of L-BPA
`varied between 290 and 400 mg/kg given at a constant
`rate over 2 h intravenously before irradiation (Table 2).
`L-BPA-F solution was infused at the BNCT facility, and
`irradiation with the neutron beam started about 45 min
`after completion of infusion.
`Blood samples for monitoring whole blood boron
`concentration were taken immediately before starting
`L-BPA-F infusion, and thereafter at about 20 min inter-
`vals during L-BPA-F infusion, following infusion, after
`treating the first portal with epithermal neutron irra-
`diation, and the last one or two samples were taken
`after completion of irradiation. The blood samples were
`analyzed for blood boron concentration using induc-
`tively coupled plasma-atomic emission spectrometry
`(ICP-AES) as described elsewhere [12]. Estimation of
`
`BPA-based BNCT performed at other BNCT facilities,
`notably at the Brookhaven National Laboratory, had
`already produced preclinical and clinical data that
`further improves safety of irradiations and formed a
`basis for further development of clinical BNCT [6,7].
`In this paper we describe the methodology used by
`the Finnish BNCT consortium in clinical BNCT trials,
`and describe shortly the first clinical results obtained
`at the Finnish BNCT facility. Since BNCT is consid-
`ered as an experimental form of radiation therapy, all
`our patients have been treated within the context of
`clinical research protocols approved by an institutional
`ethical committee, and a written informed consent was
`obtained from all patients.
`
`Patients and methods
`
`The neutron beam
`
`The neutron beam obtained from FiR 1 is moder-
`ated using Fluental™, which is composed of 69%
`aluminumfluoride, 30% aluminum, and 1% lithium
`fluoride. Circular collimator apertures of 8, 11, 14,
`17, and 20 cm in diameter are available for clinical
`use. The measured thermal (<0.5 eV), epithermal
`(0.5 eV–10 keV), and fast neutron (>10 keV) flu-
`ence rates are 8.1 × 107, 1.1 × 109, and 3.4 × 107
`neutrons/cm2/s, respectively, at the exit plane using
`a 14 cm diameter collimator at 250 kW power [8].
`The undesired fast neutron dose per epithermal flu-
`ence is 2 Gy/1013 cm−2 and the corresponding gamma
`contamination 0.5 Gy/1013 cm−2 [2]. The in-depth dose
`characteristics of the epithermal neutron beam are
`shown in Figure 2. The beam monitoring instrumen-
`tation includes three neutron sensitive fission counters
`and one gamma-sensitive ionization chamber posi-
`tioned in the neutron beam second to the moderator
`substance [9]. The instrument readings are monitored
`with a computer program and back-up hardware coun-
`ters to ensure the beam stability during irradiation.
`The main purpose of the monitoring system is to pro-
`vide a dosimetric link with the patient dose during the
`treatment. The fission chamber count rates have been
`calibrated to the induced thermal neutron fluence rate
`and to the absorbed dose rate at reference conditions
`in a tissue substitute phantom. A computational model
`for the neutron beam has been constructed, and the
`model has been verified by several measurement cam-
`paigns [10]. The model accurately produces neutron
`
`CFAD v. Anacor, IPR2015-01776, CFAD EXHIBIT 1073 - Page 3 of 12
`
`

`
`126
`
`the average whole blood boron concentration during
`irradiation was based on kinetic models [13].
`
`Radiation dose planning
`
`Conventional cranial imaging for BNCT dose plan-
`ning was done 1–3 weeks before BNCT delivery with
`a 1.5T Magnetom Vision MRI imager. Gadolinium-
`DTPA was used as a contrast agent. MRI detectable
`markers were placed on the skin before MRI to mark the
`reference points for head positioning, and their loca-
`tions were tattooed on the skin. The MRIs taken before
`craniotomy and 1–2 days after craniotomy were not
`used for dose planning, but were examined for addi-
`tional information regarding tumor localization, tumor
`volume, and presence of edema.
`The 3D Monte Carlo software packages BNCT Rtpe
`and/or SERA (INEEL/MSU, Idaho Falls/Bozeman,
`USA) were used in the BNCT dose planning. Contrast
`enhanced T1-weighted MR images were used to con-
`struct a computed 3D model of the patient’s head. The
`tissue compositions for transport computations were
`defined according to the ICRU Report 46 [14]. The
`weighted total dose (DW) was defined as the sum of
`physical dose components (Di) multiplied by weight-
`ing factors (wi) of each dose component in a tissue
`DW = wgDg + wBDB + wNDN + wfast nDfast n,
`where Dg is the gamma dose, DB the boron dose, DN
`the nitrogen dose, and Dfast n the fast neutron dose [15].
`The weighting factor for boron dose wB was taken as
`3.8 in the target and the tumor, and 1.3 in the normal
`brain. Weighting factors wN and wfast n were taken as
`3.2, and wg was considered to be 1.0 in the target, the
`tumor, and the normal brain [16,17]. The fluence-to-
`kerma conversions of the weighted nitrogen and the
`weighted fast neutron doses were calculated using a
`nitrogen concentration of 1.84 wt % and a hydrogen
`concentration of 10.57 wt %, assuming the brain tissue
`to be composed of equal proportions of the white and
`gray matter [18]. The unit for the physical dose com-
`ponents is Gy and for the weighted dose Gy (W).
`The doses in the tumor, the target volume, and in
`the sensitive tissues were computed individually as
`a function of the average boron concentration in the
`whole blood during irradiation. For the boron con-
`centration, tumor-to-whole blood ratio of 3.5 : 1 and
`the normal brain-to-whole blood ratio of 1 : 1 were
`assumed. The computational head model consisted of
`the skin, the skull, the brain, the target volume, and
`
`the tumor regions in protocol P-01. In addition to these
`structures, the sinuses were also outlined in protocol
`P-03. When computing the average brain dose, the
`entire brain and the tumor site were included in the
`computation volume. The target volume was defined
`to consist of the enhancing tumor present in MRI, the
`surrounding edema, plus a 1–2 cm margin in the brain
`tissue in three dimensions. Two fields were irradiated in
`all cases, and an attempt was made to exclude the con-
`tralateral hemisphere from the target volume whenever
`possible. Maximum doses allowed in dose planning
`were determined for different anatomical structures.
`
`Patient positioning
`
`Patient positioning simulation for irradiation was car-
`ried out one day preceding irradiation. The beam entry
`and exit coordinates were provided by the dose plan-
`ning program. The entry and exit coordinates given
`by the dose planning program were transformed in
`a Microsoft Excel program to a positioning coordi-
`nate system with the help of three detectable reference
`markers, which were placed on the patient’s skin before
`carrying out the dose planning MRI. Patient position-
`ing was performed in the treatment simulation room
`located next to the nuclear reactor. The computed beam
`exit and entry points were first localized and marked on
`the skin. After finding the optimal head position relative
`to the beam aperture, head and body vacuum immobi-
`lizers were shaped to secure maintained head and body
`position during neutron beam irradiation. The patient
`positioning system included a custom-made treatment
`coach equipped with electrical controls for the couch
`table position in three dimensions (Te-Pa Medical Oy,
`Lappeenranta, Finland), a beam aperture simulator, and
`a total of nine crosshair lasers. The crosshair laser sys-
`tem was fixed to the center of the beam aperture, and
`provided an identical coordinate system for head posi-
`tioning both in the simulation room and in the irradia-
`tion room.
`
`Irradiation and monitoring of the irradiation dose
`
`Following BPA-F infusion, the patient was placed in
`the preshaped vacuum immobilizers on the treatment
`couch. The correctness of the head position for treat-
`ment was verified using positioning lasers first in the
`simulator and then at the irradiation site immediately
`before irradiation. All treatments were given as one sin-
`gle fraction. The irradiation time of the first field ranged
`
`CFAD v. Anacor, IPR2015-01776, CFAD EXHIBIT 1073 - Page 4 of 12
`
`

`
`from 15.2 to 40.2 min (median, 29.6 min), patient repo-
`sitioning between the fields took about 20 min, and irra-
`diation of the second field lasted from 14.5 to 37.2 min
`(median, 21.5 min). Hence, the irradiation procedure
`typically lasted for about 1 h. Apertures of 11 or 14 cm
`in diameter were used in all irradiations. During neu-
`tron beam irradiation, the patient position was mon-
`itored with two television cameras. Pulse and blood
`oxygen level were monitored during irradiation. Vital
`signs were recorded before irradiation, and at 2-h inter-
`vals following irradiation for 8 h.
`The nominal irradiation time computed with a dose
`planning program was adjusted based on the whole
`blood boron concentrations measured at the reactor site
`with ICP-AES. The blood boron analysis results were
`available about 10 min after sampling. The average
`blood boron concentration during each neutron irradia-
`tion was estimated based on kinetic models and preir-
`radiation blood boron concentration data. Two kinetic
`models, an open two-compartment model and a bi-
`exponential fit are currently in use in the Finnish BNCT-
`trials (Figure 3). These models estimate the clearance
`of boron from the blood after BPA-F infusion of 290 mg
`BPA/kg body weight with accuracy of about 1 ppm or
`less during the first and second radiation fields [13].
`Recently, a more capable kinetic model was developed
`[19]. The target beam monitor counts were set based
`on the corrected irradiation times. The irradiation was
`terminated by a reactor scram when the set beam mon-
`itor counts were reached.
`Absorbed gamma doses were measured using in vivo
`thermoluminescence dosimeters (TLD) placed in the
`ipsilateral ear canal, at the fixation point ventral to the
`contralateral ear, at the base of the nose, and on the skin
`over the 7th cervical vertebra, the thyroid, the sternum,
`and on the umbilicus. Thermal neutron fluences were
`measured with Mn(n, γ ) activation foils/wires placed
`on the beam entry points, in the ipsilateral ear canal,
`and at the base of the nose.
`
`Patient follow-up
`
`After neutron beam irradiation, the patients were fol-
`lowed up at the Department of Oncology, Helsinki
`University Central Hospital
`for about 2–3 days
`for possible acute radiation-related adverse effects.
`Dexamethasone was routinely prescribed to pre-
`vent radiation-related edema, and all patients also
`received antiepileptic medication. Neurological status
`and adverse effects were recorded using structured
`
`127
`
`forms. Brain MRI examinations were scheduled to be
`performed 1, 3, 6, 9, 12, 18, and 24 months after irradi-
`ation using gadolinium-DTPA as a contrast agent, and
`clinical follow-up visits were performed at 1–3 month
`intervals during the first post-irradiation year.
`
`Protocols
`
`Protocol P-01
`
`P-01 is a prospective, nonrandomized, phase I to
`II study focusing on feasibility of giving BNCT as
`primary radiotherapy to patients with newly diag-
`nosed glioblastoma multiforme. Eighteen glioblastoma
`patients have been enrolled between May 1999 and
`December 2001. Eleven patients were male and the
`median age was 55.5 (ranging 31–67). The median time
`interval from surgery to BNCT was 31.5 days (rang-
`ing from 15 to 43 days). The inclusion and exclusion
`criteria are presented in Table 1. BPA is given intra-
`venously complexed with fructose as 30 g BPA/L aque-
`ous solution over 2 h, and the BPA-F dosage given
`
`Table 1. P-01 protocol inclusion and exclusion criteria (BNCT
`as primary treatment for glioblastoma following surgery)
`
`Inclusion criteria
`Histologically confirmed glioblastoma multiforme
`Supratentorial location
`Age 18–75
`Karnofsky’s performance status 70% or higher
`Adequate antiepileptic medication
`Written informed consent is obtained and the patient is able to
`understand the nature of the trial
`Exclusion criteria
`Radiation tolerance of the optic chiasma or the basal ganglia is
`estimated to be exceeded in dose planning, or an adequate dose
`is not considered to be achieved in the deep-seated parts of the
`target
`Less than 30% of the tumor has been removed at surgery based
`on comparison of the preoperative MRI and a postoperative
`MRI taken no longer than 72 h after craniotomy
`Over 6 week time interval from craniotomy to BNCT
`Prior cranial radiation therapy, cancer immunotherapy,
`chemotherapy or gene therapy
`Serious cardiac insufficiency, liver or renal disease, or infection
`Presence of a cardiac pacemaker, or metallic prostheses or
`implants in the head and neck area that prohibit MRI
`Pregnancy or breast feeding
`Phenylketonuria
`Dexamethason is contraindicated
`
`CFAD v. Anacor, IPR2015-01776, CFAD EXHIBIT 1073 - Page 5 of 12
`
`

`
`Open 2-compartment model
`Measured data
`2 postinfusion data points
`3 postinfusion data points
`All data points
`Irradiations
`
`0
`
`0.5
`
`1
`
`1.5
`
`2
`
`2.5
`
`3
`
`3.5
`
`4
`
`4.5
`
`5
`
`Time (h)
`
`(a)
`
`25
`
`20
`
`15
`
`10
`
`5
`
`0
`
`Boron concentration (ppm)
`
`128
`
`Biexponent fit
`Measured data
`2 postinfusion data points
`3 postinfusion data points
`All data points
`Irradiations
`
`(b)
`
`25
`
`20
`
`15
`
`10
`
`5
`
`0
`
`Boron concentration (ppm)
`
`0
`
`0.5
`
`1
`
`1.5
`
`2
`
`2.5
`
`3
`
`3.5
`
`4
`
`4.5
`
`5
`
`Time (h)
`
`Figure 3. (a) 10B concentration time-behavior based on an open 2-compartment model and blood boron concentration data from a patient
`infused with 290 mg BPA/kg body weight. The measured data points are expressed as circles. The results are obtained using data available
`before the first irradiation field (chain curve, 2 postinfusion data points used for fitting), before the second irradiation field (broken curve,
`3 postinfusion data points used for fitting), and when all data points are available for fitting (solid curve). (b) A similar analysis using a
`bi-exponential function fitting instead of the 2-compartment model. Chain curve, 2 postinfusion data points used for fitting; broken curve,
`3 postinfusion data points used for fitting; solid line, all data points available for fitting.
`
`varied between 290 and 400 mg BPA/kg body weight.
`Blood hematology and chemistry are monitored before
`irradiation, 1 day after irradiation, and at the 1- and
`3-month follow-up visits.
`
`The BNCT Rtpe dose planning program was used.
`The protocol limits the maximum average weighted
`dose to the normal brain as 7 Gy (W). The BPA dosages,
`the average weighted planning target volume (PTV)
`
`CFAD v. Anacor, IPR2015-01776, CFAD EXHIBIT 1073 - Page 6 of 12
`
`

`
`129
`
`Brain, Dave = 5 Gy (W)
`Target, Dave = 61 Gy (W)
`Tumor, Dave = 62 Gy (W)
`
`50
`
`45
`
`40
`
`35
`
`30
`
`25
`
`20
`
`15
`
`10
`
`05
`
`Fraction of volume, %
`
`0
`
`5
`
`10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85
`
`Weighted total dose, Gy (W)
`
`Figure 4. An example of a dose–volume histogram showing the
`computed weighted doses in the normal brain, the target, and the
`tumor (patient 18 in Table 2).
`
`doses, and the average weighted doses to the normal
`brain are shown in Table 2 for the 18 patients treated.
`The average PTV doses were first increased in the
`first 12 cases while keeping the BPA-F dosage con-
`stant at 290 mg BPA/kg, following which we gradu-
`ally increased the BPA-F dosage from 290 to 400 mg
`BPA/kg body weight in patients 13–18. The average
`weighted dose to the normal brain has remained at
`about 5 Gy (W). The normal brain weighted peak doses
`were less than 14 Gy (W) (Table 2), and the normal
`brain physical (unweighted) peak doses less than 11 Gy
`(Table 3). An example of a computed dose–volume his-
`togram is shown in Figure 4. The protocol may accrue
`a maximum of 20 patients.
`The BPA-F infusion was well tolerated, and BNCT-
`related acute toxicity has been acceptable. The only
`serious (Grade 3 or 4) toxicity related to BNCT con-
`sists of acute abdominal pain leading to laparotomy in
`one case. Transient dysphasia lasting for a few days
`was observed in six patients, transient amnesia in three,
`and six patients had an epileptic fit within the first week
`following BNCT. None of the patients have died during
`the first months following BNCT. The 6-month over-
`all survival is 100% (5 patients have been followed
`up for less than 6 months), and estimated 1-year sur-
`vival 61%. Most patients received further cancer treat-
`ments following recurrence. Since follow-up is still
`incomplete, overall toxicity, time to progression, and
`overall survival results will be presented in a more
`detail in a future report. An example of treatment result
`following BNCT is shown in Figure 5.
`
`Table 2. BPA dosages, and the average weighted doses in
`18 glioblastoma patients treated in the P-01 protocol
`
`Case Gender/ BPA
`age
`dosage
`(mg/kg)
`
`Normal
`Average
`Average
`brain
`normal
`planning
`target volume brain dose peak dose
`dose (range)
`(range)
`(Gy(W))
`(Gy (W))
`(Gy (W))
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`
`F/63
`F/50
`M/59
`F/60
`M/47
`M/48
`M/67
`M/66
`F/52
`F/44
`M/57
`F/49
`M/31
`M/54
`M/64
`M/60
`M/39
`F/57
`
`290
`290
`290
`290
`290
`290
`290
`290
`290
`290
`290
`290
`330
`360
`360
`360
`400
`400
`
`30 (12–43)
`36 (15–55)
`44 (16–69)
`48 (16–79)
`45 (18–65)
`49 (21–68)
`45 (18–68)
`49 (18–78)
`47 (15–72)
`54 (24–74)
`50 (24–72)
`50 (17–75)
`43 (14–75)
`45 (17–70)
`54 (28–74)
`48 (17–71)
`42 (17–64)
`61 (32–84)
`
`3 (0–9)
`5 (2–10)
`5 (1–12)
`5 (0–14)
`5 (1–13)
`5 (1–13)
`5 (1–13)
`5 (1–13)
`5 (1–13)
`6 (1–14)
`5 (0–14)
`6 (1–14)
`5 (1–13)
`5 (1–12)
`4 (1–12)
`5 (1–12)
`5 (1–12)
`5 (1–13)
`
`8.1
`9.3
`12.0
`13.5
`12.3
`12.3
`11.9
`12.4
`10.8
`12.7
`13.0
`13.7
`12.4
`11.1
`12.0
`12.1
`11.3
`12.0
`
`Table 3. The normal brain maximum (peak) physical doses
`delivered (protocol P-01)
`
`Total
`Case Physical Physical Physical Physical
`boron
`gamma
`nitrogen
`fast neutron physical
`dose
`dose
`dose
`dose
`dose
`(Gy)
`(Gy)
`(Gy)
`(Gy)
`(Gy)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`
`2.7
`3.0
`4.5
`5.1
`4.2
`4.4
`4.4
`5.1
`3.8
`4.6
`4.7
`4.9
`4.8
`4.6
`5.0
`4.9
`4.2
`5.5
`
`2.9
`3.5
`3.9
`4.4
`4.3
`4.3
`3.9
`3.7
`3.7
`4.2
`4.4
`4.7
`4.0
`3.2
`3.6
`3.8
`3.9
`3.1
`
`0.5
`0.5
`0.6
`0.7
`0.7
`0.7
`0.6
`0.6
`0.6
`0.7
`0.7
`0.7
`0.6
`0.5
`0.6
`0.6
`0.6
`0.5
`
`0.1
`0.1
`0.2
`0.2
`0.2
`0.2
`0.2
`0.2
`0.2
`0.2
`0.2
`0.2
`0.2
`0.2
`0.1
`0.1
`0.1
`0.1
`
`6.2
`7.1
`9.2
`10.4
`9.4
`9.6
`9.1
`9.6
`8.3
`9.7
`10.0
`10.5
`9.6
`8.5
`9.3
`9.4
`8.8
`9.2
`
`CFAD v. Anacor, IPR2015-01776, CFAD EXHIBIT 1073 - Page 7 of 12
`
`

`
`130
`
`Before BNCT
`
`1 month after BNCT
`
`3 months after BNCT
`
`Figure 5. An example of irradiation result in a 39-year-old man with histologically confirmed glioblastoma multiforme. Left panel: A
`transaxial MRI scan taken 10 days after brain surgery showing an enhancing tumor in the left insular lobe. Middle panel: An MRI taken
`one month following BPA-based BNCT suggesting tumor response (the patient used dexamethason 6 mg/day). Right panel: An MRI
`three months following BNCT (the patient has been without corticosteroids for about 1.5 months).
`
`Protocol P-03
`
`P-03 is a prospective, nonrandomized, phase I study.
`The main purpose of the study is to find out whether
`BPA-based BNCT is feasible in patients with recurrent
`or progressing glioblastoma who have received prior
`cranial conventional external beam radiotherapy. The
`primary end-point is treatment-related toxicity. The
`secondary end-points include progression-free survival
`(PFS), overall survival, and quality of life. The proto-
`col was opened in February 2001, and three patients
`have been treated since. The patient eligibility criteria
`are given in Table 4.
`Therapy consists of tumor biopsy or debulking
`surgery to confirm histological diagnosis of recur-
`rent/progressing glioblastoma and to remove some
`tumor tissue. Patients receive a 2-h intravenous infu-
`sion of BPA-F that delivers 290 mg BPA/kg body
`weight before neutron beam irradiation. BNCT is
`given as a single fraction usually through two por-
`tals. The brain peak dose as computed to the max-
`imum volume of 1 cm3 is limited to less than 8 Gy
`
`(W), average normal brain dose to ≤6 Gy (W), and
`the minimum planned tumor dose must be ≥17 Gy
`(W) (requires a favorable tumor location). In the
`3–6 first patients treated the normal brain peak dose
`will be limited to a maximum of 7 Gy (W) (ranging
`5.0–7.0 Gy (W)). The SERA dose planning program
`will be used. The protocol will accrue a maximum of 22
`patients.
`The BPA dosages, average weighted PTV doses,
`and average weighted doses to the normal brain are
`shown in Table 5. In the first three treated patients, the
`therapy was well tolerated, and no serious short-term
`toxicity was encountered. Two patients have died of
`progressing glioblastoma 5 and 7 months after giving
`BNCT, but the third patient is alive at 12+ months.
`The small number of patients treated precludes mak-
`ing firm conclusions, but taking into account the very
`poor outcome of patients with recurring/progressing
`glioblastoma after full-dose conventional radiotherapy
`and the relatively good tolerability of BNCT in the first
`patients, the protocol will remain open and continues
`to accrue more patients.
`
`CFAD v. Anacor, IPR2015-01776, CFAD EXHIBIT 1073 - Page 8 of 12
`
`

`
`Table 4. Protocol P-03 inclusion and exclusion criteria (recurrent
`or persisting glioblastoma following external photon irradiation)
`
`Inclusion criteria
`Histologically confirmed supratentorial glioblastoma
`Recurrent or progressing glioblastoma after surgery and
`radiotherapy
`The total prior radiation therapy dose given is 50–60 Gy
`Conventional fractionation schemes have been used
`(conventional: 1.8–2.0 Gy/day, 5 days per week, weekly dose
`9–10 Gy)
`Recurrence/progression has been confirmed by serial MRI scans
`and a biopsy, or debulking reoperation
`The WHO performance status ≤2
`WBC >2,500/mm3, platelets >75,000/mm3, serum creatinine
`<180 µmol/l
`An informed consent is obtained
`
`Exclusion criteria
`Age <18
`Glioblastoma that infiltrates the brain stem or the optic tracts
`A minimum gross tumor dose of 17 Gy (W) is not obtained in
`dose-planning
`Less than 6 months has elapsed from the last date of
`conventional photon irradiation
`Less than 4 weeks has elapsed from the last cancer
`chemotherapy dose prior to giving BNCT
`More than approximately one-third of the total brain volume has
`been within the 90% isodose
`Gliomas where the enhancing tumor volume is larger than
`two-third of the volume of one hemisphere in the MRI
`examination preceding BNCT
`More than one radiotherapy course has been given to the
`brain tumor
`Untreated congestive heart failure or renal failure
`Uncontrolled brain edema despite use of corticosteroids
`A cardiac pacemaker or unremovable metal implants present in
`the head and neck region that will interfere with MRI-based
`dose-planning
`Restlessness or inability to lie in a cast for 30–60 min
`Clinical follow-up after therapy cannot be arranged
`Pregnancy
`The patient is not able to understand the treatment options
`The patient is not willing to participate in the follow-up schedule
`
`Table 5. BPA dosages, and average weighted doses given in 3
`patients with recurrent/progressing glioblastoma (P-03 protocol)
`
`Normal
`Average
`Case Gender/ BPA-F Average
`brain
`normal
`age
`dosage
`planning
`(mg/kg)
`target volume brain dose peak dose
`dose (range)
`(range)
`(Gy(W))
`(Gy (W))
`(Gy (W))
`
`1
`2
`3
`
`M/65
`M/52
`M/42
`
`290
`290
`290
`
`29 (14–39)
`25 (8–39)
`25 (8–41)
`
`2 (0–7)
`3 (0–7)
`3 (0–8)
`
`6.5
`7.1
`7.0
`
`131
`
`Figure 6. 18F-BPA PET image of a patient with recurrent anaplas-
`tic meningeoma on the right sphenoidal wing. The tumor-to-brain
`18F-BPA uptake ratio was 2.5–3.5 suggesting potential feasibility
`of BNCT for treating this tumor.
`
`Future protocols
`
`Protocols under development include a protocol where
`BNCT is given shortly preceding stereotactically
`guided, conformal fixed-field photon radiotherapy,
`and another protocol where patients with different
`histological types of brain tumor are selected for BNCT
`based on in vivo measured uptake of 18F-labeled L-BPA
`(L[18F]FBPA) in positron emission tomography (PET)
`(Figure 6).
`
`Discussion
`
`The median survival of glioblastoma patients is less
`than 12 months with conventional therapy, which usu-
`ally consists of surgery and radiotherapy. The purpose
`of protocol P-01 was to study the safety and toler-
`ability of BPA-based BNCT using gradually escalat-
`ing doses of irradiation and BPA. The results suggest
`that BPA-based BNCT can be safely given with the
`doses used, and that the technique is feasible. How-
`ever, the small patient number treated and the short
`follow-up preclude making comparisons regarding effi-
`cacy of BPA-based BNCT with conventional radiation
`therapy. The extent and the quality of primary and sec-
`ondary surgery, patient selection, and concomitant and
`
`CFAD v. Anacor, IPR2015-01776, CFAD EXHIBIT 1073 - Page 9 of 12
`
`

`
`132
`
`subsequent other therapies also confound such compar-
`isons. Yet, the absence of serious adverse effects and
`the relatively favorable 1-year overall survival figure of
`61% warrant further study on BPA-based BNCT.
`Neither the optimal epithermal neutron irradiation
`technique nor the preferred method of delivering the
`boron carrier compound have been established, and
`refinements in either or both of these components of
`BPA-based BNCT may result in a clinical benefit. Since
`the toxicity of BPA-based BNCT has been acceptable
`when BPA doses up to 400 mg/kg has been infused
`within 2 h prior to irradiation, still higher BPA doses
`might be studied in future protocols. However, recent
`studies performed in the rat 9L gliosarcoma model sug-
`gest that longer than 2-h infusion times are needed
`to increase the boron concentration in the infiltrat-
`ing tumor cells outside the main tumor mass, other-
`wise the infiltrating tumor cells may remain under-
`dosed [20]. These findings suggest that the 2-h infu-
`sion time may not be optimal and that infusion times
`of 6 h or longer need to be addressed in other proto-
`cols (Capala et al., in t