Download Accelerated Communication Distribution of the Phosphatidylinositol

1521-009X/42/7/1110–1116$25.00
DRUG METABOLISM AND DISPOSITION
Copyright ª 2014 by The American Society for Pharmacology and Experimental Therapeutics
http://dx.doi.org/10.1124/dmd.114.057513
Drug Metab Dispos 42:1110–1116, July 2014
Accelerated Communication
Distribution of the Phosphatidylinositol 3-Kinase Inhibitors Pictilisib
(GDC-0941) and GNE-317 in U87 and GS2 Intracranial Glioblastoma
Models—Assessment by Matrix-Assisted Laser Desorption
Ionization Imaging
Laurent Salphati, Sheerin Shahidi-Latham, Cristine Quiason, Kai Barck, Merry Nishimura,
Bruno Alicke, Jodie Pang, Richard A. Carano, Alan G. Olivero, and Heidi S. Phillips
Received February 4, 2014; accepted April 22, 2014
ABSTRACT
Glioblastoma multiforme (GBM) is the most common primary
brain tumor in adults, and the limited available treatment options
have not meaningfully impacted patient survival in the past
decades. Such poor outcomes can be at least partly attributed to
the inability of most drugs tested to cross the blood-brain barrier
and reach all areas of the glioma. The objectives of these studies
were to visualize and compare by matrix-assisted laser desorption ionization (MALDI) imaging mass spectrometry the
brain and tumor distribution of the phosphatidylinositol 3-kinase
(PI3K) inhibitors pictilisib (GDC-0941, 2-(1H-indazol-4-yl)-6-(4methanesulfonyl-piperazin-1-ylmethyl)-4-morpholin-4-yl-thieno[3,2-d]
pyrimidine) and GNE-317 [5-(6-(3-methoxyoxetan-3-yl)-7-methyl-4morpholinothieno[3,2-d]pyrimidin-2-yl)pyrimidin-2-amine] in U87 and
GS2 orthotopic models of GBM, models that exhibit differing bloodbrain barrier characteristics. Following administration to tumor-bearing
mice, pictilisib was readily detected within tumors of the contrastenhancing U87 model whereas it was not located in tumors of the
nonenhancing GS2 model. In both GBM models, pictilisib was not
detected in the healthy brain. In contrast, GNE-317 was uniformly
distributed throughout the brain in the U87 and GS2 models. MALDI
imaging revealed also that the pictilisib signal varied regionally by up
to 6-fold within the U87 tumors whereas GNE-317 intratumor levels
were more homogeneous. Liquid chromatography coupled with
tandem mass spectrometry (LC-MS/MS) analyses of the nontumored
half of the brain showed pictilisib had brain-to-plasma ratios lower
than 0.03 whereas they were greater than 1 for GNE-317, in agreement
with their brain penetration properties. These results in orthotopic
models representing either the contrast-enhancing or invasive areas
of GBM clearly demonstrate the need for whole-brain distribution to
potentially achieve long-term efficacy in GBM.
Introduction
heterogeneity of GBM (Phillips et al., 2006; Bonavia et al., 2011; Little
et al., 2012; Brennan et al., 2013; Cloughesy et al., 2014), but it is also
likely caused by the inability of the drugs tested to reach their targets.
Although disruption of the BBB commonly occurs in high-grade brain
tumors (Huse and Holland, 2010) and delivery of anticancer agents into
the contrast-enhancing core of the tumor has been shown (Hofer and
Frei, 2007; Holdhoff et al., 2010), the invasive areas are shielded by an
intact blood-brain barrier (BBB) and the protective effects of efflux drug
transporters such as P-glycoprotein (Pgp) and breast cancer resistance
protein (BCRP) (de Vries et al., 2006). Thus, effective treatment of
GBM requires that these invasive areas as well as the core of the tumor
receive adequate exposure to therapeutic agents.
Mutations of key signaling proteins of the phosphatidylinositol
3-kinase (PI3K) pathway have been detected in more than 80% of
Glioblastoma multiforme (GBM) is the most common and aggressive
primary brain tumor in adults. This tumor type, diagnosed in more than
10,000 patients each year (CBTRUS, 2012), is characterized by rapid
growth and diffuse invasiveness, and presents very few treatment
options. Disease progression is controlled for only a limited time, with
median survival less than 2 years after initial diagnosis (Adamson et al.,
2009). Despite the emergence and clinical testing of various targeted
agents expected to act on pathways deregulated in GBM, few successes
or improvements have been obtained (Huang et al., 2009). Such lack
of clinical benefit may be attributed to the remarkable molecular
dx.doi.org/10.1124/dmd.114.057513.
ABBREVIATIONS: BBB, blood-brain barrier; BCRP, breast cancer resistance protein; DCE-MRI, dynamic contrast enhanced magnetic resonance
imaging; EGFR, epidermal growth factor receptor; GBM, glioblastoma multiforme; GDC-0941, pictilisib, 2-(1H-indazol-4-yl)-6-(4-methanesulfonylpiperazin-1-ylmethyl)-4-morpholin-4-yl-thieno[3,2-d]pyrimidine; GNE-317, 5-(6-(3-methoxyoxetan-3-yl)-7-methyl-4-morpholinothieno[3,2-d]pyrimidin-2-yl)pyrimidin-2-amine; LC-MS/MS, liquid chromatography coupled with tandem mass spectrometry; MALDI, matrix-assisted laser desorption
ionization; MRI, magnetic resonance imaging; MS, mass spectrometry; NEX, number of excitations; PI3K, phosphatidylinositol 3-kinase; Pgp,
P-glycoprotein; ROIs, regions of interest.
1110
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 6, 2017
Departments of Drug Metabolism and Pharmacokinetics (L.S., S.S-L., C.Q., J.P.), Biomedical Imaging (K.B., R.A.C.), Cancer
Signaling and Translational Oncology (M.N., B.A., H.S.P.), and Chemistry (A.G.O.), Genentech, Inc., South San Francisco, California
1111
MALDI Imaging of PI3K Inhibitor Distribution in Brain Tumors
Thermo Fisher Scientific (Waltham, MA) and of analytical or high-performance
liquid chromatography grade. All other chemicals and reagents were purchased
from Sigma-Aldrich (St. Louis, MO) unless otherwise specified.
Tumor Models and MRI
Fig. 1. Structures of pictilisib (A) and GNE-317 (B).
Materials and Methods
Chemicals
Pictilisib (Fig. 1A) and GNE-317 (Fig. 1B) were synthesized by Genentech,
Inc. (South San Francisco, CA). All solvents used in analytical assays were from
MALDI Imaging
Tissue Preparation. After the MRI experiments, a single oral dose of either
pictilisib or GNE-317 was administered at 150 mg/kg and 50 mg/kg,
respectively. Treatments were administered at a time that ensured that the
BBB had recovered from the surgery and tumors were expanding. The
formulation for both compounds was 0.5% methylcellulose/0.2% Tween 80
(MCT). Mice were euthanized at 1 hour after the dose via exsanguination by
perfusion under anesthesia. Brains were excised, flash frozen in liquid N2, and
stored in a 280°C freezer until analyzed. Fresh-frozen tissue sections were
obtained on a cryomicrotome (CM3050S; Leica, Buffalo Grove, IL) at 12-mm
thickness and thaw-mounted onto indium tin oxide–coated glass slides (Bruker
Daltonics, Billerica, MA). Tissue sections were analyzed by imaging MALDI
MS, providing signal intensities (and not absolute quantitation), followed by
cresyl violet staining for histologic interrogation.
Imaging MALDI Magnetic Spectrometry. A 40 mg/ml solution of 2,5dihydroxybenzoic acid (Sigma-Aldrich) was prepared in methanol:water (70:30
TABLE 1
Plasma concentration, brain concentration, and brain-to-plasma ratio after oral administration of GNE-317 (50 mg/kg) or
pictilisib (150 mg/kg) to U87 tumor-bearing mice
Results reported as mean 6 S.D. (n = 3).
Normal Brain
Compound
Time
Plasma
Brain
mM
h
GNE-317
Pictilisib
1
6
1
6
2.01
1.86
7.77
5.26
6
6
6
6
Brain-to-Plasma Ratio
Free Brain-to-Free Plasma Ratio
mM
0.48
0.93
0.26
0.33
2.50
2.02
0.16
0.0669
6
6
6
6
0.24
1.13
0.11
0.0060
1.32
1.10
0.021
0.0127
6
6
6
6
0.48
0.21
0.014
0.0011
0.48
0.398
0.0059
0.00353
6
6
6
6
0.17
0.075
0.0038
0.00032
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 6, 2017
GBM (Network TCGAR, 2008), leading to persistent activation of the
pathway. Epidermal growth factor receptor (EGFR) amplification and/
or mutation, mutation of the PI3K catalytic and regulatory units, and
loss of PTEN (phosphatase and tensin homolog) protein are observed
in 45%, 10%, and 50% of GBM cases, respectively (Akhavan et al.,
2010). Thus, inhibition of PI3K signaling represents an attractive
therapeutic approach against GBM. However, most agents tested that
target this pathway, such as erlotinib (EGFR), gefitinib (EGFR),
lapatinib (EGFR), pictilisib (GDC-0941 [2-(1H-indazol-4-yl)-6-(4methanesulfonyl-piperazin-1-ylmethyl)-4-morpholin-4-yl-thieno[3,2-d]
pyrimidine]; PI3K), or everolimus (mTor), have been shown to
be substrates of Pgp and/or BCRP (Chu et al., 2009; Polli et al.,
2009; Agarwal et al., 2010; Salphati et al., 2010; de Vries et al.,
2012), limiting their brain penetration and ability to reach areas
beyond the tumor core. Previous studies showed efficacy of the
PI3K/mTor inhibitor GNE-317 [5-(6-(3-methoxyoxetan-3-yl)-7methyl-4-morpholinothieno[3,2-d]pyrimidin-2-yl)pyrimidin-2-amine]
in an intracranial GBM tumor model presenting an intact BBB, whereas
pictilisib displayed activity only in the U87 model, which presents a
compromised BBB (Salphati et al., 2012). However, although differences in brain and tumor penetration were likely responsible for the
contrasts in efficacy, compound tumor distribution or concentrations
were not evidenced.
The goals of our studies were to visualize the spatial distribution of the
PI3K inhibitors pictilisib and GNE-317 in the U87 and GS2 orthotopic
models and surrounding brain by matrix-assisted laser desorption
ionization (MALDI) imaging and to show the impact of different BBB
statuses, assessed by magnetic resonance imaging (MRI), on the tumor
penetration of these two compounds.
All studies conducted were approved by the Institutional Animal Care and
Use Committee at Genentech, Inc.
Two human glioma models were used in the distribution studies: U-87.luc
(U87) glioblastoma cancer cells, derived from the U-87 MG cells (American
Type Culture Collection, Manassas, VA), and the GS2 glioblastoma cells
(Gunther et al., 2008). Six female nude mice (Charles River Laboratories,
Hollister, CA) were implanted with each of the glioma cell types. The U87 (250
K) and GS2 (100 K) tumor cells were injected via stereotactic surgery into the
right striatum at a volume of 3–5 ml. Nineteen to 21 days after implantation,
one animal in each tumor group was selected for MRI to confirm the growth of
the tumor and assess the BBB integrity. MRI was performed on a Varian 9.4T
MRI system (Varian, Palo Alto, CA) with a 30-mm quadrature volume coil.
During the imaging, animals were kept under anesthesia with 2% isoflurane in air.
Body temperature was continuously monitored using a rectal probe and was
maintained at 37°C by a heated-airflow system regulated by in-house LabVIEW
controller software (National Instruments, Austin, TX).
A T2-weighted fast spin echo, multislice (FSEMS) sequence was used to
detect lesions by MRI. We acquired 12 axial 0.5 mm-thick slices with a 20 20
mm field of view (FOV), and 128 128 matrix, zero-filled to 256 256 images;
TR (repetition time) = 3500 milliseconds, TE (echo time) = 10 milliseconds, ETL
(echo train length) = 8, k-zero = 4, NEX (number of excitations) = 8. The BBB
integrity was evaluated by dynamic contrast enhanced MRI (DCE-MRI).
Precontrast three-dimensional gradient echo (3DGE) datasets were acquired at
2° and 10° flip angles, TR (repetition time) = 8.3 milliseconds, TE (echo time) =
1.1 milliseconds, NEX = 4, FOV (field of view) = 20 20 8 mm, matrix = 64 64 16. A 50-ml bolus injection of Gd-based Gadodiamide (Omniscan; GE
Healthcare, Piscataway, NJ) contrast agent was injected via a tail vein catheter
after collection of the precontrast images. Postcontrast three-dimensional gradient
echo (3DGE) images were then acquired approximately every 10 seconds for
30 minutes (10° flip angle, NEX = 1).
1112
Salphati et al.
TABLE 2
Plasma concentration, brain concentration, and brain-to-plasma ratio after oral administration of GNE-317 (50 mg/kg) or
pictilisib (150 mg/kg) to GS2 tumor-bearing mice
Results reported as mean 6 S.D. (n = 3).
Normal Brain
Compound
Time
Plasma
Brain
mM
h
GNE-317
Pictilisib
1
6
1
6
2.45
1.61
5.51
7.25
6
6
6
6
Brain-to-Plasma Ratio
Free Brain-to-Free Plasma Ratio
mM
0.45
0.35
1.21
0.73
3.79
1.87
0.150
0.178
6
6
6
6
1.09
0.38
0.057
0.061
6
6
6
6
0.21
0.02
0.0089
0.0082
0.555
0.422
0.00752
0.00684
6
6
6
6
0.076
0.007
0.00245
0.00229
under continuous accumulation of selected ions (CASI) windows optimized for
each analyte (50–125 Da). Laser intensity and the number of shots were
optimized for sensitivity of each analyte (;500–700 shots) with ion detection
collected over the mass range of m/z 150–900. Drug images were generated
based on accurate mass of each compound (pictilisib m/z 514.1690; GNE-317
m/z 415.1547) using FlexImaging v4.0 64-bit (Bruker Daltonics, Billerica, MA)
with a mass tolerance of 64 mDa and normalized to internal standard response.
Fig. 2. Brain distribution of pictilisib and GNE-317
in the U87 orthotopic model of GBM. Distribution
of pictilisib (A) and GNE-317 (B) in U87 tumors
after oral administration of 150 mg/kg and 50 mg/kg,
respectively. Localization of the tumors by cresyl
violet staining and drug distribution in MALDI MS
images are presented. For animals 1 and 4, T2weighted images of U87 intracranial tumors are
displayed to provide the location of the tumor regions
in MR images. Dynamic contrast-enhanced MR
imaging (DCE-MRI; 5 minutes after Gd injection)
showing intense contrast in the tumor confirm the
disruption of the BBB in this model.
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 6, 2017
v/v). A stable-labeled internal standard, [D8]GDC-0941, was spiked into the
MALDI matrix solution at 2 mM before deposition onto the tissue sections.
Matrix solution was homogenously spray-coated onto the tissue using a HTX
TM-Sprayer (HTX Technologies, Chapel Hill, NC). Matrix-coated tissue
sections were transferred to the MALDI mass spectrometer (MS) (SolariX 7T
FT-ICR, Bruker Daltonics, Bremen, Germany) for imaging analysis. Imaging
data were collected at 100 mm pixel resolution in positive ionization mode,
1.53
1.17
0.0271
0.0247
MALDI Imaging of PI3K Inhibitor Distribution in Brain Tumors
tumor tissue. Average drug signal from ROIs representing areas of the highest
and lowest drug intensities within the tumor tissue were compared with the
average drug signal from an arbitrarily selected nontumor region (NT) of the
brain tissue.
Quantitation of Drugs in Brain and Plasma
At time of brain harvesting for MALDI imaging analysis, blood samples
(0.2 ml) were also collected via cardiac puncture from each mouse and deposited
into tubes containing dipotassium ethylenediaminetetraacetic acid (K2 EDTA) as
an anticoagulant. After mixing the blood with K2 EDTA, we stored the samples
on ice; within 1 hour of collection, they were centrifuged for 5 minutes at 2000g
and 2–8°C. Plasma was collected and stored at 280°C until analysis.
Total concentrations of pictilisib and GNE-317 in both plasma and brain
tissue (from nontumor frontal cortex) were determined by liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). Brains were
processed and analyzed by LC-MS/MS after completion of the MALDI
analyses. Mobile phases were prepared in 0.1% formic acid in water (v/v; mobile
phase A) and 0.1% formic acid in acetonitrile (v/v; mobile phase B). Weights
from isolated brain tissue were recorded before homogenization in 4 volumes of
water. After a protein precipitation with 100% acetonitrile, supernatants from
brain and plasma extractions were injected onto a Phenomenex (Torrance, CA)
Fig. 3. Brain distribution of pictilisib and GNE317 in the GS2 orthotopic model of GBM.
Distribution of pictilisib (A) and GNE-317 (B) in
GS2 tumors after oral administration of 150 mg/kg
and 50 mg/kg, respectively. Localization of the
tumors by cresyl violet staining and drug
distribution in MALDI MS images are presented.
For animals 7 and 10, T2-weighted images of
GS2 intracranial tumors are displayed to provide
the location of the tumor regions in MR images.
DCE-MRI (5 minutes after Gd injection) showing
no contrast enhancement indicate that the BBB is
intact in this model.
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 6, 2017
Threshold intensities were set according to average internal standard response
across the two imaging experiments, which allowed for the direct comparison
of the images obtained with the two compounds.
Histology. Upon completion of the imaging experiments, the matrix coating
was removed by rinsing the glass slide in 100% methanol for 30 seconds or
until the entire matrix was visibly removed. Tissue sections were stained using
a freshly prepared 0.5% cresyl violet staining solution (Chaurand et al., 2004)
by submerging the glass slide for 30 seconds, then they were rinsed for an
additional 30 seconds in two cycles of 100% ethanol. Microscope images were
obtained on an Olympus BX51 (Olympus, Tokyo, Japan) at 10 magnification and
stitched using MicroSuite Analytical v3.0 software (Olympus). Subsequently, stained
images were coregistered to the optical images in FlexImaging for visualization
and annotation of tumor and nontumor regions for the drug images.
Intratumor Distribution. To assess intratumor distribution, imaging MALDI
MS data from the U87 and GS2 tumor models were processed in FlexImaging
v4.0 64 bit (Bruker Daltonics) across the mass range of m/z 400–525. The
cresyl violet stained microscope images were coregistered to the extracted
ion images corresponding to the exact mass of each drug, with a mass tolerance
of 64 mDa. Image threshold values were independently adjusted to obtain the
best visualization of dynamic range for each drug signal as it was detected
across the brain tissue. Regions of interest (ROIs) were selected based on the
MALDI MS ion image to assess the variation of drug distribution within the
1113
1114
Salphati et al.
Kinetex Phenyl-Hexyl column (30 2.1 mm, 2.6 mm particle size) using
a Nexera SIL 30ACMP autosampler (Shimadzu Corp., Kyoto, Japan) coupled to
a 1290 Infinity pump (Agilent Technologies, Santa Clara, CA) running a 48second gradient of 3% to 97% mobile phase B at 1.2 ml/min.
Analysis was performed on an ABSciex QTrap 5500 mass spectrometer
(Applied Biosystems, Foster City, CA) in positive ion mode monitoring the
transitions m/z 514.1→338.1 and m/z 415.1→385.1 for pictilisib and GNE-317,
respectively. Plasma and brain concentrations were determined against a 1/x
weighted quadratic fit curve range of 0.005 mM to 40 mM. Free brain
concentrations of pictilisib and GNE-317 were determined as previously
described (Salphati et al., 2010).
Results and Discussion
Fig. 4. Distribution of pictilisib in U87 tumor (A, animal 2) and GNE-317 in U87 (B, animal 6) and GS2 (C, animal 12) tumors and drug distribution ratios. ROIs (A–F)
were selected based on the MALDI MS ion image to assess the variation of drug distribution within the tumor tissue. Average drug signal from ROIs representing areas of the
highest and lowest drug intensities within the tumor tissue were compared with the average drug signal from an arbitrarily selected nontumor region (NT) of the brain tissue.
Signal ratios of tumor:nontumor response are reported as a fold-change in drug distribution (blue, lowest ratio; red, highest ratio). The mean value corresponds to the ratio of
the average signal from the entire tumor to the nontumor region.
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 6, 2017
Pictilisib, a PI3K inhibitor currently being evaluated in phase II
clinical trials, is a substrate for the efflux transporters Pgp and BCRP,
which limit its brain penetration (Salphati et al., 2010). In contrast,
GNE-317, a PI3K/mTor inhibitor, is able to cross the BBB (Heffron
et al., 2012). In our present study, we evaluated by MALDI imaging
the distribution of these two compounds in brains of mice bearing
intracranial tumors. Two GBM models were selected: U87 and GS2
glioma cells. The U87 cell line is a well-characterized and widely used
glioma model (Weller et al., 1998; Clark et al., 2010). These cells
grow adherently in tissue culture and, when implanted as xenografts,
develop well-delineated tumors with impaired BBB. In contrast, the
GS2 cells grow in vitro as neurospheres and recapitulate the diffusive
infiltration and mostly intact BBB of the invading regions of human
GBM (Salphati et al., 2012).
Pictilisib and GNE-317 were administered as a single oral dose, at
150 and 50 mg/kg, respectively, to mice bearing either tumor. Doses
were selected based on previous pharmacokinetic and efficacy studies.
Plasma concentrations of the two compounds measured 1 and 6 hours
after the dose (Tables 1 and 2) were consistent with our previous
studies (Salphati et al., 2010; Salphati et al., 2012). Brain levels
obtained from the nontumored hemisphere after tissue homogenization
confirmed the brain penetration potential of each compound, with total
brain-to-plasma ratios lower than 0.03 for pictilisib and greater than
1 for GNE-317 (Tables 1 and 2). These ratios did not markedly differ
between 1 and 6 hours, indicating that the distribution was not time
dependent and suggesting that analysis of intratumor distribution
1 hour after the dose would be representative of later time points.
Although they are informative, these data do not provide insight
into the spatial distribution of the compounds within the brain or
tumor, specifically. Further understanding of brain drug delivery was
obtained with MALDI imaging, which clearly showed the distribution
of the compounds within the brain and/or the tumors. In the U87
model, pictilisib had levels close to the limit of detection in most of the
brain whereas it was more readily detected specifically in the tumor
(especially in animal 2: Fig. 2A and Fig. 4A), as annotated by cresyl
violet staining and MRI (Fig. 2A).
The highly localized distribution of pictilisib in the U87 tumor, with
a signal approximately 13-fold higher than in the healthy part of the
brain (Fig. 4A), can be explained by the impairment of the BBB in this
GBM model, highlighted by the marked contrast enhancement after
gadolinium injection (Fig. 2A, animal 1; Fig. 2B, animal 4), which
allows even poorly brain-penetrant compounds such as pictilisib to
leave the blood capillaries and distribute in the tumor. In addition,
differences in the signal among the three animals bearing the U87
tumors (Fig. 2A) underscore the variability in exposure in the same
MALDI Imaging of PI3K Inhibitor Distribution in Brain Tumors
concentrations in the invasive areas in addition to the core of the
tumor.
In summary, our studies provide visualization of the difference in
distribution for brain-penetrant or nonpenetrant compounds in two
intracranial GBM models that represent different regions and features
of human GBM. Our findings also support what had been hypothesized regarding the regional drug delivery in glioma (Agarwal
et al., 2011) and confirm that meaningful clinical benefit, and reliable
assessment of pharmacodynamics (biological) hypotheses, will most
likely be achieved in GBM (or brain metastases) only with compounds
able to distribute throughout the whole brain.
Authorship Contributions
Participated in research design: Salphati, Shahidi-Latham, Quiason,
Nishimura, Alicke, Pang, Olivero, Phillips.
Conducted experiments: Barck, Nishimura, Alicke.
Contributed new reagents or analytic tools: Shahidi-Latham, Quiason,
Barck, Carano.
Performed data analysis: Salphati, Shahidi-Latham, Quiason, Barck, Pang,
Carano, Phillips.
Wrote or contributed to the writing of the manuscript: Salphati, ShahidiLatham, Barck, Nishimura, Phillips.
References
Adamson C, Kanu OO, Mehta AI, Di C, Lin N, Mattox AK, and Bigner DD (2009) Glioblastoma
multiforme: a review of where we have been and where we are going. Expert Opin Investig
Drugs 18:1061–1083.
Agarwal S, Manchanda P, Vogelbaum MA, Ohlfest JR, and Elmquist WF (2013) Function of the
blood-brain barrier and restriction of drug delivery to invasive glioma cells: findings in an
orthotopic rat xenograft model of glioma. Drug Metab Dispos 41:33–39.
Agarwal S, Sane R, Gallardo JL, Ohlfest JR, and Elmquist WF (2010) Distribution of gefitinib to
the brain is limited by P-glycoprotein (ABCB1) and breast cancer resistance protein (ABCG2)mediated active efflux. J Pharmacol Exp Ther 334:147–155.
Agarwal S, Sane R, Oberoi R, Ohlfest JR, and Elmquist WF (2011) Delivery of molecularly
targeted therapy to malignant glioma, a disease of the whole brain. Expert Rev Mol Med 13:
e17.
Akhavan D, Cloughesy TF, and Mischel PS (2010) mTOR signaling in glioblastoma: lessons
learned from bench to bedside. Neuro-oncol 12:882–889.
Bonavia R, Inda MM, Cavenee WK, and Furnari FB (2011) Heterogeneity maintenance in
glioblastoma: a social network. Cancer Res 71:4055–4060.
Brennan CW, Verhaak RG, McKenna A, Campos B, Noushmehr H, Salama SR, Zheng S,
Chakravarty D, Sanborn JZ, and Berman SH, et al.; TCGA Research Network (2013) The
somatic genomic landscape of glioblastoma. Cell 155:462–477.
CBTRUS(2012) CBTRUS Statistical Report: Primary Brain and Central Nervous System Tumors
Diagnosed in the United States in 2004–2008, Central Brain Tumor Registry of the United
States, Hillsdale, IL, www.cbtrus.org.
Chaurand P, Schwartz SA, Billheimer D, Xu BJ, Crecelius A, and Caprioli RM (2004) Integrating
histology and imaging mass spectrometry. Anal Chem 76:1145–1155.
Chu C, Abbara C, Noël-Hudson MS, Thomas-Bourgneuf L, Gonin P, Farinotti R,
and Bonhomme-Faivre L (2009) Disposition of everolimus in mdr1a-/1b- mice and after a pretreatment of lapatinib in Swiss mice. Biochem Pharmacol 77:1629–1634.
Clark MJ, Homer N, O’Connor BD, Chen Z, Eskin A, Lee H, Merriman B, and Nelson SF (2010)
U87MG decoded: the genomic sequence of a cytogenetically aberrant human cancer cell line.
PLoS Genet 6:e1000832.
Cloughesy TF, Cavenee WK, and Mischel PS (2014) Glioblastoma: from molecular pathology to
targeted treatment. Annu Rev Pathol 9:1–25.
Daniel WA, Wójcikowski J, and Pałucha A (2001) Intracellular distribution of psychotropic drugs
in the grey and white matter of the brain: the role of lysosomal trapping. Br J Pharmacol 134:
807–814.
de Vries NA, Beijnen JH, Boogerd W, and van Tellingen O (2006) Blood-brain barrier and
chemotherapeutic treatment of brain tumors. Expert Rev Neurother 6:1199–1209.
de Vries NA, Buckle T, Zhao J, Beijnen JH, Schellens JH, and van Tellingen O (2012) Restricted
brain penetration of the tyrosine kinase inhibitor erlotinib due to the drug transporters P-gp and
BCRP. Invest New Drugs 30:443–449.
Fine RL, Chen J, Balmaceda C, Bruce JN, Huang M, Desai M, Sisti MB, McKhann GM,
Goodman RR, and Bertino JS, Jr, et al. (2006) Randomized study of paclitaxel and tamoxifen
deposition into human brain tumors: implications for the treatment of metastatic brain tumors.
Clin Cancer Res 12:5770–5776.
Günther HS, Schmidt NO, Phillips HS, Kemming D, Kharbanda S, Soriano R, Modrusan Z,
Meissner H, Westphal M, and Lamszus K (2008) Glioblastoma-derived stem cell-enriched
cultures form distinct subgroups according to molecular and phenotypic criteria. Oncogene 27:
2897–2909.
Heffron TP, Salphati L, Alicke B, Cheong J, Dotson J, Edgar K, Goldsmith R, Gould SE, Lee LB,
and Lesnick JD, et al. (2012) The design and identification of brain penetrant inhibitors of
phosphoinositide 3-kinase a. J Med Chem 55:8007–8020.
Hofer S and Frei K (2007) Gefitinib concentrations in human glioblastoma tissue. J Neurooncol
82:175–176.
Holdhoff M, Supko JG, Gallia GL, Hann CL, Bonekamp D, Ye X, Cao B, Olivi A, and Grossman
SA (2010) Intratumoral concentrations of imatinib after oral administration in patients with
glioblastoma multiforme. J Neurooncol 97:241–245.
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 6, 2017
tumor model. It is worth also noting that differences in intensity,
corresponding to up to 6-fold changes in pictilisib signal within the
U87 tumor, can be observed (Fig. 4A). This range of intratumor signal
is most likely reflective of the variable degrees of BBB disruption
within the tumor, leading to heterogeneous distribution.
Such variability in concentrations with efflux transporter substrates
can also be expected between and within patients, depending on the
degree and localization of the BBB disruption. Thus, optimal compound concentrations may not be achieved throughout the tumor,
which is likely to limit efficacy. This finding is consistent with the
variability observed with radiolabeled paclitaxel, doxorubicin (Lockman
et al., 2010), and lapatinib (Taskar et al., 2012) in a model of brain
metastasis of breast cancer, with up to a 10-fold range in concentrations between and within the tumor metastases. The U87 model,
with its compromised BBB, represents what may occur in the contrastenhancing region(s) of a human brain tumor. Indeed, specific localization of erlotinib in the contrast-enhancing area of a brain
metastasis in a patient with non-small cell lung cancer was demonstrated
by Weber et al. (2011). Similarly, high concentrations of compounds,
even those known to be substrates of efflux transporters, have been
reported in resected human brain tumors (primary or metastases), with
lower levels in surrounding tissues (Fine et al., 2006; Pitz et al., 2011).
The lack of meaningful long-term clinical benefit even in cases
when a high compound concentration was measured in the tumor can
be explained, at least in part, by the absence of compound distribution
beyond the regions with disrupted BBB. In preclinical models, lower
concentrations of erlotinib (Agarwal et al., 2013) have been measured
in the tissues surrounding and in areas distant from the core of the
glioma. These findings relied on dissecting the tumors and adjacent
tissues from the rest of the brain, which did not allow the assessment
of intratumor heterogeneity or whole-brain distribution.
In brains implanted with the GS2 tumors, which present an intact
BBB, as shown by the absence of contrast-enhancement (Fig. 3), postcontrast, pictilisib was not detected in any area (Fig. 3A), in agreement
with the poor brain penetration of this agent. The brain distribution of
GNE-317 was strikingly different from that of pictilisib. Regardless of
the tumor model, and the BBB status (as assessed by gadolinium
enhancement), GNE-317 was uniformly distributed throughout the
brain and tumors (Fig. 2B, and Fig. 3B). Distribution appeared
consistent between mice (Fig. 2B, and Fig. 3B) and more homogenous
than that of pictilisib in the U87 tumor (Fig. 4B), with similar average
signal in the tumor and the normal brain regions selected (Fig. 4B),
and a 2.7-fold range in signal among the regions assessed within the
tumor (Fig. 4B). In the GS2 model, the GNE-317 average tumor signal
appeared 2.5-fold lower than in the nontumor region selected whereas
the intensity varied by about 4-fold within the tumor (Fig. 4C). The
weaker signal in the GS2 tumor versus the healthy brain as well as the
higher variability in distribution within the tumor compared with the
U87 tumor may be explained by differences in tumor vascularization,
which could impact GNE-317 distribution. Images suggest also
different compound levels in white versus gray matter as previously
reported with brain penetrant compounds (Daniel et al., 2001; Li et al.,
2009).
These results clearly explained why efficacy was observed with
GNE-317 but not pictilisib in the GS2 model (Salphati et al., 2012).
The GS2 model, with its intact BBB and diffuse growth, can be
viewed as representative of the invasive areas of human GBM.
Therefore, the stark difference in brain and tumor distribution between
GNE-317 and pictilisib, which led to marked differences in efficacy,
may translate into the clinical setting. Pictilisib may be able to
distribute to the core of the tumor but not to the invasive glioma cells.
GNE-317, and not pictilisib, may be able to achieve efficacious
1115
1116
Salphati et al.
Huang TT, Sarkaria SM, Cloughesy TF, and Mischel PS (2009) Targeted therapy for malignant
glioma patients: lessons learned and the road ahead. Neurotherapeutics 6:500–512.
Huse JT and Holland EC (2010) Targeting brain cancer: advances in the molecular pathology of
malignant glioma and medulloblastoma. Nat Rev Cancer 10:319–331.
Li F, Hsieh Y, Kang L, Sondey C, Lachowicz J, and Korfmacher WA (2009) MALDI-tandem
mass spectrometry imaging of astemizole and its primary metabolite in rat brain sections.
Bioanalysis 1:299–307.
Little SE, Popov S, Jury A, Bax DA, Doey L, Al-Sarraj S, Jurgensmeier JM, and Jones C (2012)
Receptor tyrosine kinase genes amplified in glioblastoma exhibit a mutual exclusivity in
variable proportions reflective of individual tumor heterogeneity. Cancer Res 72:1614–1620.
Lockman PR, Mittapalli RK, Taskar KS, Rudraraju V, Gril B, Bohn KA, Adkins CE, Roberts A,
Thorsheim HR, and Gaasch JA, et al. (2010) Heterogeneous blood-tumor barrier permeability
determines drug efficacy in experimental brain metastases of breast cancer. Clin Cancer Res
16:5664–5678.
Network TCGAR; Cancer Genome Atlas Research Network (2008) Comprehensive genomic
characterization defines human glioblastoma genes and core pathways. Nature 455:1061–1068.
Phillips HS, Kharbanda S, Chen R, Forrest WF, Soriano RH, Wu TD, Misra A, Nigro JM,
Colman H, and Soroceanu L, et al. (2006) Molecular subclasses of high-grade glioma predict
prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis.
Cancer Cell 9:157–173.
Pitz MW, Desai A, Grossman SA, and Blakeley JO (2011) Tissue concentration of systemically
administered antineoplastic agents in human brain tumors. J Neurooncol 104:629–638.
Polli JW, Olson KL, Chism JP, John-Williams LS, Yeager RL, Woodard SM, Otto V, Castellino
S, and Demby VE (2009) An unexpected synergist role of P-glycoprotein and breast cancer
resistance protein on the central nervous system penetration of the tyrosine kinase inhibitor
lapatinib (N-3-chloro-4-[(3-fluorobenzyl)oxy]phenyl-6-[5-([2-(methylsulfonyl)ethyl]aminomethyl)2-furyl]-4-quinazolinamine; GW572016). Drug Metab Dispos 37:439–442.
Salphati L, Heffron TP, Alicke B, Nishimura M, Barck K, Carano RA, Cheong J, Edgar KA,
Greve J, and Kharbanda S, et al. (2012) Targeting the PI3K pathway in the brain—efficacy of
a PI3K inhibitor optimized to cross the blood-brain barrier. Clin Cancer Res 18:6239–6248.
Salphati L, Lee LB, Pang J, Plise EG, and Zhang X (2010) Role of P-glycoprotein and breast
cancer resistance protein-1 in the brain penetration and brain pharmacodynamic activity of the
novel phosphatidylinositol 3-kinase inhibitor GDC-0941. Drug Metab Dispos 38:1422–1426.
Taskar KS, Rudraraju V, Mittapalli RK, Samala R, Thorsheim HR, Lockman J, Gril B, Hua E,
Palmieri D, and Polli JW, et al. (2012) Lapatinib distribution in HER2 overexpressing experimental brain metastases of breast cancer. Pharm Res 29:770–781.
Weber B, Winterdahl M, Memon A, Sorensen BS, Keiding S, Sorensen L, Nexo E,
and Meldgaard P (2011) Erlotinib accumulation in brain metastases from non-small cell lung
cancer: visualization by positron emission tomography in a patient harboring a mutation in the
epidermal growth factor receptor. J Thorac Oncol 6:1287–1289.
Weller M, Rieger J, Grimmel C, Van Meir EG, De Tribolet N, Krajewski S, Reed JC, von
Deimling A, and Dichgans J (1998) Predicting chemoresistance in human malignant glioma
cells: the role of molecular genetic analyses. Int J Cancer 79:640–644.
Address correspondence to: Dr. Laurent Salphati, Department of Drug
Metabolism and Pharmacokinetics, MS 412A, Genentech, Inc., 1 DNA Way,
South San Francisco, CA 94080. E-mail: [email protected]
Downloaded from dmd.aspetjournals.org at ASPET Journals on May 6, 2017