MR Studies of Glioblastoma Models Treated with Dual … · Tumor Registry of the United States for...

Companion Diagnostics and Cancer Biomarkers

MR Studies of Glioblastoma Models Treated withDual PI3K/mTOR Inhibitor and Temozolomide:Metabolic Changes Are Associated withEnhanced SurvivalMarinaRadoul1,MyriamM.Chaumeil1, Pia Eriksson1, AlanS.Wang1, Joanna J. Phillips2,3, andSabrina M. Ronen1,2

Abstract

The current standard of care for glioblastoma (GBM) issurgical resection, radiotherapy, and treatment with temozo-lomide (TMZ). However, resistance to current therapies andrecurrence are common. To improve survival, agents that targetthe PI3K signaling pathway, which is activated in approximate-ly 88% of GBM, are currently in clinical trials. A challenge withsuch therapies is that tumor shrinkage is not always observed.New imaging methods are therefore needed to monitorresponse to therapy and predict survival. The goal of this studywas to determine whether hyperpolarized 13C magnetic reso-nance spectroscopic imaging (MRSI) and 1H magnetic reso-nance spectroscopy (MRS) can be used to monitor response tothe second-generation dual PI3K/mTOR inhibitor voxtalisib(XL765, SAR245409), alone or in combination with TMZ. We

investigated GS-2 and U87-MG GBM orthotopic tumors inmice, and used MRI, hyperpolarized 13C MRSI, and 1H MRSto monitor the effects of treatment. In our study, 1H MRS couldnot predict tumor response to therapy. However, in both ourmodels, we observed a significantly lower hyperpolarized lac-tate-to-pyruvate ratio in animals treated with voxtalisib, TMZ,or combination therapy, when compared with controls. Thismetabolic alteration was observed prior to MRI-detectablechanges in tumor size, was consistent with drug action, andwas associated with enhanced animal survival. Our findingsconfirm the potential translational value of the hyperpolarizedlactate-to-pyruvate ratio as a biomarker for noninvasively asses-sing the effects of emerging therapies for patients with GBM.Mol Cancer Ther; 15(5); 1113–22. �2016 AACR.

IntroductionGlioblastoma (GBM) is one of the most aggressive and fatal

types ofmalignant gliomawith a very poor prognosis regardless ofhow early the tumor is detected. According to the Central BrainTumor Registry of the United States for 2006–2010, GBMaccounts for more than 50% of all newly diagnosed glioma cases(1). Current standard of care for GBM is safe surgical resection,followed by radiotherapy and temozolomide (TMZ) treatment(2, 3). However, diffuse infiltration of tumor cells into eloquentregions of the normal brain precludes full surgical resection. Inaddition, development of resistance to radiotherapy and TMZ iscommon (4, 5), and the heterogeneous biology of GBM furthercomplicates therapy (6). Thus, despite harsh anticancer treat-ments, GBM tumors always recur. Outcomes can vary with tumor

size and location, but unfortunately average survival is still lessthan 2 years (1, 7). In light of this dismal prognosis, recent workhas focused on the development of alternative and complemen-tary treatment strategies.

Data from The Cancer Genome Atlas shows that approximately88% of GBM have activated the PI3K pathway (8). This centralsignalingpathway is involved in essential cellular functions, such asgrowth, migration, proliferation, metabolism, and survival. ThePI3K pathway can be activated at different steps of the signalingcascade, most notably by loss of the PTEN protein, and by ampli-fication and/or mutation of the EGFR, which occur in 40% and50% of GBM cases, respectively (8). Accordingly, several drugs thatinhibit signaling downstream of these mutations are currently inphase I and II clinical trials, either in combination with other FDA-approved chemotherapeutic agents, such TMZ, or as single agents(9–13). Importantly, whereas first-generation PI3K pathway inhi-bitors target mostly the downstream node of mTOR, the observa-tion of feedback loops that result in enhanced PI3K signaling hasled to the development of second-generation drugs (14–17). Thesetarget both mTOR and PI3K and generally overcome the negativefeedback (18, 19).

Current clinical evaluation of response to treatment for GBM ismost commonly based on tumor size as detected by MRI (20).However, in the case of PI3K pathway inhibitors, treatmentoften induces tumor stasis rather than shrinkage, limiting theutility of anatomic imaging (21). Alternative imaging methodsthat can probe for drug–target engagement and assess responseare therefore critically needed to improve timely clinical patientmanagement.

1Department of Radiology and Biomedical Imaging, University ofCalifornia San Francisco, San Francisco, California. 2Brain TumorResearchCenter,UniversityofCalifornia SanFrancisco, SanFrancisco,California. 3NeuropathologyDivision,DepartmentofPathology,UCSFSchool of Medicine, UCSF Medical Center, San Francisco, California.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Sabrina M. Ronen, University of California, San Fran-cisco, Box 2532 UCSF Mission Bay Campus, Byers Hall, Rm BH303E, 1700 4thStreet, San Francisco, CA 94158. Phone: 415-514-4839; Fax: 415-514-4839;E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-15-0769

�2016 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org 1113

on May 31, 2020. © 2016 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 16, 2016; DOI: 10.1158/1535-7163.MCT-15-0769

http://mct.aacrjournals.org/

One possible approach is the imaging of metabolism. Thevalue of 1H magnetic resonance spectroscopy (MRS) in detect-ing the metabolic alterations characteristic of GBM has beenclearly demonstrated in the clinic (22–24). Levels of MRS-detectable total choline-containing metabolites (tCho) are ele-vated in tumors when compared with normal brain, whereaslevels of N-acetyl aspartate (NAA) are reduced. Interestingly,elevated tCho and phosphocholine (PC) levels have also beenlinked to activation of PI3K signaling, and inhibition of thissignaling pathway has resulted in a drop in PC levels in severalcancer types, including GBM (25–28). Based on these previousfindings, tCho levels could serve to assess response to PI3Kpathway inhibitors.

An alternate approach formonitoringmodulations of the PI3Kpathway is the recently developed hyperpolarized 13C magneticresonance spectroscopic imaging (MRSI) approach. Hyperpolar-ization improves the signal-to-noise ratio for 13C-labeled sub-strates by over 10,000-fold, and thus enables rapid noninvasivein vivo imaging of the metabolism of hyperpolarized substrates(29–31). Previous studies from our group have shown that thelevel of hyperpolarized [1-13C]-lactate produced from exogenoushyperpolarized [1-13C]-pyruvate decreases in several GBM celllines following inhibition of the PI3K pathway either at the levelof PI3K or at the level ofmTOR. A drop in hyperpolarized [1-13C]-lactate was also observed in vivo following mTOR inhibition in arat GBM model (26, 32, 33). These metabolic observations werelinked to drug action via the decreased expression of lactatedehydrogenase (LDH-A), which is controlled by PI3K signalingand catalyzes the pyruvate-to-lactate conversion (32). 13C MRSIwas also used to probe the effect of TMZ on hyperpolarized[1-13C]-lactate production. A drop in hyperpolarized [1-13C]-lactate levels was shown to precede tumor shrinkage in GBM invivo, mediated by a drop in intracellular lactate levels (34, 35).Importantly, the first-in-human clinical study using hyperpolar-ized [1-13C]-pyruvate in prostate cancer patients showed notoxicity and was highly sensitive in detecting elevated hyperpo-larized [1-13C]-lactate levels in regions of cancer (36). The hyper-polarized 13C MRSI approach could therefore also serve forclinical assessment of PI3K inhibitors, either alone or in combi-nation with TMZ.

However, none of the previous MRS studies have investigat-ed the impact in GBM of novel second-generation inhibitorsthat target both PI3K and mTOR, and that are now in clinicaltrials. Furthermore, no previous studies have determinedwhether changes in 1H MRS and hyperpolarized 13C MRSI areassociated with increased survival following treatment. Thegoal of this study was therefore to use 1H MRS and hyperpo-larized 13C MRSI to monitor the effect of a dual PI3K/mTORinhibitor, either alone or in combination with TMZ, in twoorthotopic brain tumor models. We focused on the dual PI3K/mTOR inhibitor, voxtalisib (SAR245409, XL765), which is apan-class I PI3K, mTORC1, and mTORC2 inhibitor (15, 16),and investigated its impact on two GBM models that exhibitactivated PI3K signaling via PTEN deletion, GS-2 and U87-MG,as orthotopic tumors in mice. Surprisingly, we found that tCholevels were not altered by treatment. In contrast, in both of ourmodels, the hyperpolarized lactate-to-pyruvate ratio (Lac/Pyr)as detected by 13C MRSI was significantly lower followingtreatment. The metabolic changes were associated withenhanced animal survival and generally preceded detectablechange in tumor size following treatment with voxtalisib either

alone or in combination with TMZ. Our findings highlight thevalue of hyperpolarized 13C MRSI as a potentially translationalmethod for noninvasive assessment of the effect of novelsecond-generation dual PI3K inhibitors.

Materials and MethodsCell culture

GS-2 andU87-MGhumanGBM cells (26, 32)were supplied bythe UCSF Brain Tumor Research Center Preclinical TherapeuticsCore in 2009 and 2010, respectively. Cells were routinely finger-printed by the UCSF Genomics Core Facility using SNP (lastfingerprinting August 28, 2014) and maintained in culture lessthan 6 months. Cells were cultured in DMEM (GIBCO) supple-mented with 10% FBS, 2 mmol/L L-glutamine, 100 units/mLpenicillin, and 100 mg/mL streptomycin in 5% CO2 at 37�C.

DrugsVoxtalisib (SAR245409, XL765) was kindly provided by

Sanofi (Exelixis). Voxtalisib was prepared by sonication in10 mmol/L HCl. TMZ (Sigma-Aldrich) was dissolved inORA-plus (Perrigo). All drugs were prepared right beforeadministration.

Animal models and study designAll studies were performed under the UCSF Institutional Ani-

mal Care andUse Committee approval. Six- to 7-week-old femaleathymic nu/nu mice (20–25 g) were intracranially injected withapproximately 3 � 105 GS-2 or U87-MG cells. Tumor size wasevaluatedusingMRI.Once tumors reached2 to 3mmindiameter,a baseline set of MR studies was performed (see below). This timepoint was considered day zero (D0). Mice were then randomizedinto four treatment groups and treated per os (p.o.) with: (i) 30mg/kg voxtalisib twice daily; (ii) 5 mg/kg TMZ daily; (iii) 30 mg/kg voxtalisib twice daily and 5mg/kg TMZ daily (TMZwas given 1hour after the first dose of voxtalisib); (iv) 10 mmol/L HCl daily(for controls).MR studieswere then repeated every 2 to 4days andcontinued either until the animal had to be sacrificed or until thetumor was no longer detectable. At the end of the MR studies, thebrainwas excised and divided into twohalves. Half the tumor andcontralateral normal-appearing brain were fixed in formalin forhistologic analysis. The second half was snap-frozen in liquidnitrogen and stored at �80�C until 1H high-resolution magicangle spinning (HR-MAS) MR studies. To obtain tissue for his-tologic and HR-MAS studies when treatments resulted in com-plete tumor shrinkage, additional studies were performed andanimals were sacrificed when the tumor had shrunk to its originalsize at D0. In those cases, animal survival for the Kaplan–Meiersurvival curvewas chosen as the time point atwhich tumorwas nolonger detectable.

In vivo MR studiesAll studies were performed on a 14.1T vertical MR system

(Agilent Technologies), equipped with single channel 1H coil forT2-weighted imaging and 1HMRS, or with dual-tune 1H-13C coilfor hyperpolarized 13C MRSI.

Anatomic MR imagingAxial T2-weighted images were used to evaluate tumor sizes

and were recorded using a multislice spin-echo sequencewith the following parameters: time to echo (TE), 20 ms;

Radoul et al.

Mol Cancer Ther; 15(5) May 2016 Molecular Cancer Therapeutics1114




repetition time (TR), 1,200 ms; field of view (FOV), 30 � 30mm2; matrix, 256 � 256; slice thickness, 1.8 mm; number ofaverages (NA), 2. Tumor size was determined by manuallycontouring the tumor area in each axial slice and calculatingthe total volume as a sum of the areas multiplied by slicethickness using in-house MR software (37).

Hyperpolarized 13C MR spectroscopic imaging[1-13C]-pyruvic acid was prepared as previously described

(32), and 300 mL injected intravenously through a tail-veincatheter over 12 seconds. Spectra were then acquired asdescribed below.

First, a preliminary dynamic study in a small number ofuntreated animals (n ¼ 3/model) was performed to determinethe time course of hyperpolarized [1-13C]-pyruvate deliveryand maximum hyperpolarized [1-13C]-lactate production. Forthis, a two-dimensional (2D) chemical shift imaging (CSI)dynamic sequence was used over the course of 52.5 secondsfollowing hyperpolarized [1-13C]-pyruvate injection with thefollowing parameters: TE, 1.2 ms; TR, 60 ms; flip angle (FA),10; matrix, 128 � 7 � 7; spectral width, 2,500 Hz; FOV, 30 �30 mm2; 4 mm slice thickness; scan time, 3.5 seconds per timepoint; 15 time points. Data were then processed, and peakintegrals of hyperpolarized [1-13C]-pyruvate and hyperpolar-ized [1-13C]-lactate were quantified within the tumor voxel anda normal brain voxel using the SIVIC software (38). Hyperpo-larized [1-13C]-lactate integrals were normalized to noise andto maximum hyperpolarized [1-13C]-pyruvate signal. The aver-age time points at which hyperpolarized [1-13C]-lactate wasmaximal in normal brain and in tumor voxels were thendetermined (�17 seconds for both tumor and normal brainin both models; see Results). Subsequent data were thereforerecorded using a 2D-CSI pulse sequence at approximately 17seconds after injection of hyperpolarized [1-13C]-pyruvateusing: TE, 0.41 ms; TR, 66 ms; FA, 10; matrix, 256 � 16 �16; spectral width, 4,223 Hz; FOV, 20 � 20 mm2; 4 mm slicethickness. Data were processed using SIVIC (38). A tumorvoxel was chosen after evaluation of all slices such that itcontained >80% of tumor tissue, avoiding necrotic and inho-mogeneous regions. Peak integrals corresponding to hyperpo-larized [1-13C]-pyruvate and [1-13C]-lactate in the tumor voxelwere used to calculate Lac=PyrTumor

Di , where Di is the day oftreatment (i ¼ 0, 1. . .n) and further normalized toLac=PyrContralateralDi obtained from contralateral brain such that:

Lac=PyrDi ¼ Lac=PyrTumorDi

Lac=PyrContralateralDi

:

1H MR spectroscopy1H MRS spectra were acquired from a 2 � 2 � 4 mm3 voxel

placed in the center of the tumor using a point resolvedspectroscopy (PRESS) sequence with: TE, 20 ms; TR, 4,000 ms;NA, 256–512; 4,096 points; spectral width 10,000 Hz. Spectrawere analyzed using QUEST in jMRUI using a basis set withalanine (Ala), choline (Cho), PC, glycerophosphocholine(GPC), creatine (Cr), phosphocreatine (PCr), glutamate (Glu),glutamine (Gln), glutathione (GSH), glycine (Gly), lactate(Lac), myo-inositol (mI), NAA, and taurine (Tau; ref. 39).Because Glu and Gln, Cr and PCr, and Lac and lipids overlap,these metabolites were quantified as Glx, Cr, and LacþLipids,

respectively. The integral of total choline (sum of Cho, PC, andGPC) on Di was determined and normalized to the total signalðtSignalÞDi:

ðtChoÞNormalizedDi ¼ ðtChoÞDi

tSignalð ÞDi:

1H HR-MAS spectroscopyTissue samples were weighed and 17 � 5 mg placed into a

zirconia rotor, where 6 mL of D2O with 0.75 w/% 2,2,3,3-D4-3-(trimethylsilyl)propionic acid were added as chemical shift ref-erence. All spectra were acquired on Agilent 500 MHz spectrom-eter at 1�C. Samples were spun at 2,250 Hz and Carr–Purcell–Meiboom–Gill sequence usedwith: TE, 144ms (train of 162 180�

hard pulses 0.888 ms apart) and 512 transients. Spectra wereanalyzed using MestReNova (MestreLab Research), and peakintegrals were normalized to total signal (40).

ImmunohistochemistryAfter the MR study, tissues were fixed in formalin for 24 hours,

then dehydrated with ethanol, and embedded in paraplast Pluswax (McCormick Scientific). Tumor tissue sections were probedwith antibodies for: LDH-A (Cell Signaling Technology), carbonicanhydrase IX (CAIX; Novus Biologicals), phospho-S6 ribosomalprotein (Ser240/244; Cell Signaling Technology), and phospho-Chk1 (Ser345; Cell Signaling Technology) to inform on LDH-Aexpression, HIF-1 expression, PI3K signaling, and TMZ action,respectively. The stained slides were imaged using Nikon EclipseTi-E motorized inverted microscope.

Statistical analysisResults represent mean� SD. Two-tailed unpaired Student t test

was used to determine the statistical significance of differences, andP value � 0.05 was considered significant. Kaplan–Meier survivalcurves with the log-rank test were used to compare survival.

ResultsPreliminary optimization and validation studies

Previous hyperpolarized studies to monitor response to ther-apywere performed on a 3T clinical scanner in rats and probed theeffect of the mTOR inhibitor everolimus in GS-2 tumors and theeffect of TMZ in U87-MG tumors (32, 34, 35). We therefore firstwanted to perform a small-scale study in which we optimized ourdata acquisition conditions for mice studies at 14.1T.

Tumors were detected using T2-weighted anatomical MRI(Fig. 1A and B). We then used a 13C dynamic 2D-CSI sequenceto assess the kinetics of hyperpolarized [1-13C]-pyruvate deliveryand hyperpolarized [1-13C]-lactate production in GS-2 and U87-MG tumors as well as in normal brain. Our findings for lactatebuild-up and decay are illustrated in Fig. 1C and D. We deter-mined that maximum hyperpolarized [1-13C]-lactate productionwas reached at 17.5 � 0 seconds in normal brain and 17.5 � 3.5seconds in tumor in the GS-2 model (n ¼ 3), and at 17.5 � 0seconds in normal brain and 17.5 � 4.0 seconds in tumor in theU87-MGmodel (n¼3). Basedon thesefinding,weperformedoursubsequent investigations using a 2D-CSI sequence at the singletime point of maximum hyperpolarized [1-13C]-lactate produc-tion (�17 seconds) in order to achieve the best possible sensitivityand spatial resolution.

Next, we wanted to confirm that previous findings in rats withregard to the effect of treatment on lactate production (32, 34, 35)

Hyperpolarized 13C MRSI-Detectable Changes with GBM Treatment

www.aacrjournals.org Mol Cancer Ther; 15(5) May 2016 1115




also hold inmice.We therefore treatedGS-2–bearingmice (n¼ 5)with everolimus, and U87-MG–bearing animals (n ¼ 4) withTMZ, and compared our findings with control tumors (n ¼ 5 foreach tumormodel). We confirmed that treatment of GS-2 tumorswith everolimus for 9�2days (Fig. 1E) and treatment ofU87-MGtumorswith TMZ for 6� 1 days (Fig. 1F) resulted in a significantlylower level of hyperpolarized Lac/Pyr when compared with con-trols (P<0.05 for CTRL vs. everolimus inGS-2 tumors, P<0.03 forCTRL vs. TMZ in U87-MG tumors). These findings recapitulatedthe previous observations in rats on a 3T clinical scanner, andpaved the way for our studies monitoring the dual PI3K/mTORinhibitor in mice at 14.1T (32, 34, 35).

Effect of treatment in the GS-2 modelIn vivo hyperpolarized 13C MRSI. The GS-2 xenografts reached asize of 0.1 � 0.01 cm3, sufficient for hyperpolarized 13C MRSIstudies, at 38� 9 days after implantation. Figure 2A exhibits axialT2-weighted images and hyperpolarized 13C MRSI spectra fromtumor voxels from each treatment group recorded at D0, priorto treatment, and at D9 � 2 of treatment. Average data illus-trating the temporal evolution of tumor sizes in the differenttreatment groups are shown in Fig. 2B. Whereas administrationof voxtalisib (n ¼ 5) did not significantly affect tumor growth,administration of combination voxtalisib/TMZ (n ¼ 5) andTMZ alone (n ¼ 6) led to tumor shrinkage that was significantby D14 of treatment (P < 0.05) when compared with control

animals (n ¼ 5). Next, we focused on the time interval of D9 �2, prior to any significant change in tumor size when comparingtreated animals with controls (Fig. 2C). As illustrated in Fig. 2D,treatment led to a significantly lower level of hyperpolarizedLac/Pyr at this time point for every treatment group whencompared with controls (P � 0.05 for all treatments).Kaplan–Meier survival curves (Fig. 2E) demonstrate the signif-icantly improved survival of mice treated with voxtalisib (n¼ 5,log-rank P < 0.001), voxtalisib/TMZ combination (n ¼ 5, log-rank P < 0.001), or TMZ (n ¼ 6, log-rank P < 0.001) comparedwith the controls (n ¼ 5). Lower hyperpolarized Lac/Pyrdetected in treated tumors at D9 � 2 was therefore associatedwith improved survival, whereas tumor size was unchanged atthis time, and could not inform on response.

Immunohistochemistry. Previous studies have shown that follow-ing treatment with PI3K pathway inhibitors, lower hyperpolar-ized lactate production is mediated by a reduction in the level ofHIF-1, which leads to a reduction in the expression of LDH-A, theenzyme that catalyzes the pyruvate-to-lactate conversion (32,33, 41). It has also been shown that following TMZ treatment,lower hyperpolarized lactate production is mediated by anincrease in phospho-Chk1 levels, which leads to a drop in intra-cellular lactate levels and thus hyperpolarized lactate production(34, 35). In order to confirm that our observations were mech-anistically linked to PI3K pathway inhibition and to TMZ action

-0.05

0

0.05

0.1

0.15

0.2

0.25

0.3

38.531.524.517.510.53.5

Lact

ate (

AU)

Time (s)

Normalbrain LAC

Tumor LAC

0.5

1.0

1.5

2.0

2.5

3.0

TMZCTRL

Lac/P

yr (A

U)

0.50

1.00

1.50

2.00

EverolimusCTRL

Lac/P

yr(A

U)

C D

FE

* *

GS-2

GS-2

A B U87-MGGS-2

Lac PyrLac Pyr

U87-MG

U87-MG

-0.05

0

0.05

0.1

0.15

0.2

0.25

0.3

38.531.524.517.510.53.5

Lact

ate (

AU)

Time (s)

Normalbrain LAC

Tumor LAC

Figure 1.GS-2 (left) and U87-MG (right)models: A and B, T2-weighted imagesand corresponding hyperpolarized13C MR spectra from tumor voxels.C and D, temporal evolution ofhyperpolarized [1-13C]-lactate in thetumor and in normal brain voxels.E and F, average hyperpolarizedLac/Pyr detected in tumor voxelin GS-2 tumors treated witheverolimus at D9 � 2 and inU87-MG tumors treated withTMZ at D6 � 1. �, P < 0.05.

Radoul et al.





in a similar manner, we performed immunohistochemical (IHC)analyses on our tumors resected at the end of the MR studies (Fig.3). The expression of CA-IX is controlled by HIF-1, and thereforestaining for CA-IX informed on HIF-1 levels. Our findings show adrop in CA-IX, as well as a drop in LDH-A expression, followingvoxtalisib treatment, either alone or with TMZ. Staining forphospho-S6, which is downstream of PI3K and mTOR, alsodropped following voxtalisib exposure, confirming inhibition ofsignaling via the PI3K pathway. Following treatment with TMZ,either alone or in combination with voxtalisib, an increase inphospho-Chk1 was observed. Our findings are thus consistentwith previous work, linking drug action to the metabolic altera-tions observed by MRS in GS-2 tumors.

1H MR spectroscopy. We also wanted to determine whether tChocan serve as a biomarker for tumor response to therapy with thedual PI3K/mTOR inhibitor, as previously observed in vivowith thePI3K inhibitor PX-866, and in cells with the PI3K inhibitorLY294002 (26, 28). We therefore acquired 1H MR spectra fromthe tumor voxel. Figure 4A exhibits axial T2-weighed images ofmice from the different treatment groups and their correspondingtumor spectra acquired on D9 � 2 of treatment. Figure 4Billustrates the average levels of tCho. Unexpectedly, tCho levelsdid not exhibit a significant change as a result of treatment with

voxtalisib (n ¼ 4, P ¼ 0.34). However, treatment with combina-tion voxtalisib/TMZ (n ¼ 3, P ¼ 0.06) showed a trend toward adecrease, and treatment with TMZ alone resulted in a significantlylower level of tCho by 33% (n ¼ 3, P < 0.05) compared withcontrols (n ¼ 5). No changes were detected in any other meta-bolites in the 1HMRS spectrumnor inmetabolite ratios, includingAla, Cr, GSH, NAA, m-Ins, Tau, Gly, Glx, LacþLipid, NAA/Cr, andCho/Cr (Supplementary Table S1).

To confirm our in vivo findings, we also analyzed frozenresected tumor samples using 1H HR-MAS, a method thatprovides higher spectral resolution and thus improved metab-olite quantification. Figure 4C shows the effect of treatment oncholine-containing metabolites. The HR-MAS data (Fig. 4D)confirmed the in vivo findings for all treatment groups. Levels oftCho did not change significantly as a result of treatment witheither voxtalisib (n ¼ 4, P ¼ 0.59) or combination voxtalisib/TMZ (n ¼ 4, P ¼ 0.64) when compared with controls (n ¼ 4),but in the case of treatment with TMZ, tCho was significantlylower by 31% (n ¼ 4, P < 0.05). When considering themetabolites that comprise the tCho peak, no significantchanges could be detected in any individual metabolite fol-lowing any of the treatments. Furthermore, consistent with ourin vivo findings, no other metabolites or metabolite ratioschanged with treatment (Supplementary Table S2).

0

50

100

150

200

250

300

350

4035302520151050

Tum

or si

ze (%

D0)

Time of treatment (days)

CTRL VoxtalisibVoxtalisib/TMZ TMZ

Lac Pyr

D9±2

CTRL Voxtalisib TMZD0

Lac Pyr

0

0.2

0.4

0.6

0.8

1

4035302520151050

Surv

ival p

roba

bilit

y



B EC

A

* ***

*

0.5

1.0

1.5

2.0

CTRL

Voxt

alisib

Voxt

alisib

/TM

Z TMZ

Lac/P

yr (A

U)

50

150

250

CTRL

Voxt

alisib

Voxt

alisib

/TM

Z TMZ

Tum

or si

ze (%

D0)

*D

Voxtalisib/TMZPyrLac Lac Pyr

Figure 2.GS-2 model: A, axial T2-weighted images and corresponding hyperpolarized 13C MRSI spectra from the tumor voxel at D0 (top) and D9� 2 of treatment (bottom).The tumor regions are contoured with a white line. B, temporal evolution of average tumor size as a percentage of pretreatment value. C, averagetumor size detected at D9 � 2. D, average Lac/Pyr detected at D9 � 2. E, Kaplan–Meier survival curves comparing control and treatment groups. �, P < 0.05.






Effect of treatment in the U87-MG modelTo further confirm our findings, we also investigated a second

GBM model, the U87-MG. Despite the same number of injectedcells, U87-MG tumors grew faster than GS-2 and reached a size of0.09 � 0.03 cm3, sufficient for MRS studies, by 18 � 3 days afterimplantation. We then performed similar MRS studies to thosedescribed above for the GS-2 model.

In vivo hyperpolarized 13CMRSI.Axial T2-weighted images ofmicefrom each group of treatment are shown in Fig. 5A. Images andcorresponding hyperpolarized 13C MRSI spectra were recorded atD0, prior to onset of treatment and at D6� 1 of treatment. Figure5B illustrates the temporal evolution of tumor size and shows thattreatment with voxtalisib (n ¼ 5) led to a rapid slowdown oftumor growth that was significant compared with controls (n¼ 5,P < 0.05) by D5 � 1. Treatment with TMZ (n ¼ 4) and combi-nation voxtalisib/TMZ (n ¼ 5) resulted in a slowdown of tumorgrowth that was significant by D8 � 1 of treatment and wasfollowed by frank tumor shrinkage that began aroundD12� 1 oftreatment. Focusing on D6 � 1, Fig. 5C illustrates average tumorsize. At this time point, a significant decrease in tumor size wasdetected as a result of treatment with voxtalisib, whereas changesin tumor size in the groups treated with combination voxtalisib/TMZ or TMZ were not yet significant compared with controls.However, hyperpolarized Lac/Pyr (Fig. 5D)was lower atD6�1 inall treatment groups when compared with controls (P < 0.05 forall treatments). Average survival is illustrated using the Kaplan–Meier survival curves (Fig. 5E) and shows a significant increase in

animal survival (log-rank P < 0.001) in all treatment groups. As inthe case of the GS-2 model, the early change in hyperpolarizedLac/Pyr observed at D6 � 1 of treatment was associated withenhanced rodent survival in the U87-MG model.

1H HR-MAS MR spectroscopy. Our first tumor model (GS-2) onlyshowed limited changes in tCho or any other metabolite. Wetherefore restricted our investigation of the U87-MGmodel to thehigher resolution ex vivo1H HR-MAS studies. Figure 6A illustratesthe 1H HR-MAS data and the effect of treatment on choline-containing metabolites. Average tCho levels (Fig. 6B) show verycomparable values among all treatment groups with no signifi-cant differences when compared with controls (n ¼ 3 for alltreatment group, P ¼ 0.63 for voxtalisib, P ¼ 0.52 for voxtali-sib/TMZ, and P ¼ 0.51 for TMZ). When considering the meta-bolites that comprise the tCho peak, no significant changes couldbe detected in any metabolite, and none of the other 1H MRS-detectable metabolites or metabolite ratios changed with treat-ment in the U87-MG model (Supplementary Table S3).

DiscussionThe vast majority of GBM tumors have activated their PI3K

signaling pathway (8). PI3K inhibitors are therefore in clinicaltrials for treatment of GBM, either alone or in combination withthe standard-of-care DNA-damaging agent TMZ (34, 35). How-ever, monitoring response to such therapies can be challengingbecause tumor shrinkage is not always observed (21). The goal of

CA-9 LDH-A Phospho-S6 CT

RL

Voxt

alisib

Vo

xtali

sib/T

MZ

TMZ

Phospho-Chk1

Figure 3.GS-2 model: IHC showing theexpression of CA-IX, LDH-A, phospho-S6, and phospho-Chk1 for eachtreatment group at the end of study.Bar, 20 mm.

Radoul et al.





this study was therefore to expand upon our previous cell andin vivo studies (26, 32, 33, 35) anddetermine the clinical relevanceof 1H MRS-detectable choline-containing metabolite levels and13C MRSI-detectable hyperpolarized Lac/Pyr, as early biomarkersof drug–target engagement and possible indicators of enhancedsurvival following treatment with a previously unexplored sec-ond-generation dual PI3K/mTOR inhibitor in GBM.

We studied twoorthotopic humanGBMmodel, GS-2 andU87-MG, in mice. Both of these models are PTEN deficient, and, assuch, both have an activated PI3K signaling pathway and modelthe majority of GBM tumors. Tumors in the mouse brain cannotgrow to a size that allows for more than one voxel to be placedwithin the tumor regionwithout substantial partial volume effectsfrom the surrounding normal parenchyma. As a result, anyheterogeneity in tumor response could not be detected in ourstudies. Another possible limitation of our study is that it wasperformed using a 2D-CSI sequence at the time point of maxi-mum lactate production common to both tumor and normalbrain. This allowed for better signal-to-noise compared withdynamic data. However, we cannot rule out that the time pointfor maximum lactate differs during treatment from the valueobserved in both tumor and normal brain. Nonetheless, we wereable to confirm in mice previous observations in rats showing adrop in hyperpolarized lactate production following everolimusand TMZ treatment (32, 35). Our findings are thus likely notspecies dependent.

Our studies were designed to mimic the clinic by investigatingnot only the dual PI3K/mTOR inhibitor, but also combinationtreatment with the standard-of-care TMZ. Furthermore, our stud-ies used clinically relevant dosing regimens and oral administra-tion (16, 42). Interestingly, response to voxtalisib was quitedifferent in both of our models. The U87-MG model respondedby tumor shrinkage. In contrast, the size of GS-2 tumors remainedindistinguishable from controls. A previous study in GBM39showed that voxtalisib, either alone or in combination with TMZ,leads to a reduction in tumor size as measured by biolumines-cence (43). The reasons for such differential effects on tumorgrowth remain to be determined. Nonetheless, in both of ourmodels, the MRS observations were similar and were associatedwith enhanced survival.

Our previouswork had shown that treatmentwith inhibitors ofthe PI3K pathway at the level of PI3K or mTOR alone resulted in1H and 31P MRS-detectable changes in tCho and PC both in cellsand in vivo, in GBM, and other cancers (26, 28). Surprisinglyhowever, in both of our models, our findings did not indicate asignificant change in tCho as a result of treatment with the dualPI3K/mTOR inhibitor. Previous reports investigating the effects ofBEZ235 and PI-103, two other dual PI3K/mTOR inhibitors,showed either a drop or an increase in choline-containing meta-bolites (44–46). Further studies are therefore required to under-stand these discrepancies, which are likely associated with thecomplexity of the signaling pathways that control choline

0.00

0.05

0.10

0.15

0.20

CTRL

Voxt

alisib

Voxt

alisib

/TMZ TM

Z

tCho

(AU)

GPC PC Cho

3.04.0 ppm1.52.02.53.5

0.00

0.05

0.10

0.15

0.20

CTRL

Voxt

alisib

Voxt

alisib

/TMZ TM

Z

tCho

(AU)

CTRL

Cho

CrNAA

Glx

LacLip.

D7Voxtalisib

Cho

Cr NAALacLip.

D7

D7TMZ

Cho

CrNAA Lac

Lip.

Voxtalisib/TMZCho

CrNAA

LacLip.

D9

A B

Cho

PCGPC

LacAla

GlnGlu, Gln

Gly

Taum-Ins

Cr/PCrCr, PCr

m-InsLac

m-InsGlx Glu

GSH

NAA

*Cho

GPCPC

ChoPC

GPC

Cho

GPC

PCCho

GPCPC

*

C DCTRL Voxtalisib Voxtalisib/TM TMZ

Figure 4.GS-2 model: A, axial T2-weighted images and corresponding 1H spectra of the tumor voxel from each treatment group. B, average tCho levels at D9 � 2.C, representative ex vivo1H HR-MAS spectrum of a biopsy from control tumor. The insets zoom in on choline-containingmetabolites, illustrate each treatment group.D, comparison of tCho levels and individual choline-containing metabolites. Error bars, SD of tCho.






metabolism (25). Nonetheless, our findings point to the chal-lenges of using choline metabolites as general biomarkers ofresponse to inhibitors of the PI3K pathway.

In contrast, in both our models, lower levels of hyperpolar-ized Lac/Pyr were observed following drug exposure in everytreatment group. We cannot rule out that the change in hyper-polarized pyruvate metabolism is a nonspecific effect associat-ed with response to therapy; similar metabolic effects have beenreported with starvation-induced autophagy and radiotherapy(47, 48). It is also possible that voxtalisib affects pyruvatemetabolism via MAPK signaling or via MYC, either downstreamof the PI3K pathway or via nonspecific drug effects (49, 50).Nonetheless, we did find that in the GS-2 model, voxtalisibtreatment leads to lower phospho-S6 levels, reduced HIF-1activity (evidenced by lower CA-IX), and lower expression ofLDH-A. These data confirm reduced PI3K signaling and providea mechanistic link between reduced signaling and our meta-bolic observations. Our findings are also identical to previousobservations in GBM, breast, and prostate cancer models (32,33, 41). Similarly, we observed an increase in phospho-Chk1 in

TMZ-treated tumors, linking the molecular effect of TMZ withthe lower level of hyperpolarized lactate production, as previ-ously observed in GBM (34, 35).

The metabolic effects observed in our hyperpolarized 13C MRSstudy were detected within approximately 9 days of treatment inthe case of the GS-2 model, and within approximately 6 days oftreatment in the case of the U87-MGmodel, time points at whichchanges in tumor size were generally not detectable by MRI. Assuch, our data are consistent with previous studies, showing thathyperpolarized Lac/Pyr can serve as an early indicator of drug–target engagement, and demonstrate that the hyperpolarized 13Capproach can also be used to monitor the effect of voxtalisib,possibly other dual PI3K/mTOR inhibitors, and combinationtherapy with TMZ. Our studies also show that reduced Lac/Pyris associated with enhanced survival following treatment, inde-pendent of tumor size. Indeed, although in most cases treatmenteventually affected tumor growth, in the case of GS-2 tumors,voxtalisib did not lead to a detectable change in tumor size.Nonetheless, voxtalisib-treated mice lived statistically significant-ly longer than controls.

0

200

400

600

800

1,000

181614121086420

Tum

or si

ze (%

D0)



A

C EB

D

PyrLac

D6±1

CTRL Voxtalisib Voxtalisib/TMZ TMZD0

PyrLac Lac Pyr Lac Pyr

*

**

*

0

0.2

0.4

0.6

0.8

1

181614121086420Su

rviva

l pro

babi

lity



* * **

0.50

1.50

2.50

3.50

CTRL

Voxt

alisib

Voxt

alisib

/TMZ TM

Z

Lac/P

yr (A

U)

100

200

300

400

CTRL

Voxt

alisib

Voxt

alisib

/TMZ TM

Z

Tum

or si

ze (%

D0)

*

Figure 5.U87-MG model: A, anatomical axial T2-weighted images and corresponding hyperpolarized 13C MRSI spectra from the tumor voxel at D0 (top) and D6 � 1 oftreatment (bottom). Tumor regions are contoured in a white dashed line. B, temporal evolution of average tumor size. C, average tumor size detected at D6� 1 oftreatment. D, average hyperpolarized Lac/Pyr detected at D6� 1 of treatment. E, Kaplan–Meier survival curves comparing survival between control and treatmentgroups.

Radoul et al.





The recently completedfirst-in-man imaging studydemonstrat-ed the safety of hyperpolarized [1-13C]-pyruvate as an agent fornoninvasive characterization of metabolism in patients, anddemonstrated the value of the hyperpolarized methodology forclinical studies (36). The approach used in our study couldtherefore be implemented in the clinic to aid in monitoring drugaction and assessing likely response to PI3K pathway inhibitors.Such a noninvasive imaging approach would assist the treatmentdecision-making process at an early time point allowing for theadvancement of precision medicine that is tailored to the indi-vidual patient.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: M. Radoul, S.M. RonenDevelopment of methodology: M. Radoul, S.M. RonenAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): M. Radoul, M.M. Chaumeil, P. Eriksson, A.S. Wang,J.J. Phillips

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): M. Radoul, M.M. Chaumeil, S.M. RonenWriting, review, and/or revision of the manuscript: M. Radoul, M.M.Chaumeil, J.J. Phillips, S.M. RonenAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): M. Radoul, M.M. Chaumeil, P. Eriksson,S.M. RonenStudy supervision: S.M. Ronen

AcknowledgmentsThe authors acknowledge M. VanCriekinge, S. Subramaniam, J. Crane,

B. Olson, A. Elkhaled, and R. Bok for their technical support, help with SIVIC,and animal handling, as well as support from the hyperpolarized MRI Tech-nology Resource Center (P41EB013598, to D.B. Vigneron).

Grant SupportThis study was supported by NIH grant R01CA154915 (to S.M. Ronen).The costs of publication of this articlewere defrayed inpart by the payment of

page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received September 18, 2015; revised January 13, 2016; accepted February 2,2016; published OnlineFirst February 16, 2016.

References1. Ostrom QT, Gittleman H, Liao P, Rouse C, Chen Y, Dowling J, et al.

CBTRUS statistical report: Primary brain and central nervous systemtumors diagnosed in the United States in 2006–2010. Neuro Oncol2013;15.

2. Stupp R,MasonWP, van den Bent MJ. Radiotherapy plus concomitant andadjuvant temozolomide for glioblastoma. NEngl JMed2005;352:987–96.

3. Sengupta S, Marrinan J, Frishman C, Sampath P. Impact of temozolomideon immune response during malignant glioma chemotherapy. Clin DevImmunol 2012;2012:831090.

4. Johannessen TC, Bjerkvig R. Molecular mechanisms of temozolomideresistance in glioblastoma multiforme. Expert Rev Anticancer Ther2012;12:635–42.

5. StuppR,HegiME,MasonWP, van denBentMJ, TaphoorndMJB, Janzer RC,et al. Effects of radiotherapywith concomitant and adjuvant temozolomideversus radiotherapy alone on survival in glioblastoma in a randomised

phase III study: 5-year analysis of the EORTC-NCIC trial. Lancet Oncol2009;10:459–66.

6. Kesari S. Understanding glioblastoma tumor biology: The potential toimprove current diagnosis and treatments. Semin Oncol 2011;38:Suppl 4:S2–10.

7. Park CK, Lee SH, Kim TM, Choi SH, Park SH, Heo DS, et al. The value oftemozolomide in combination with radiotherapy during standardtreatment for newly diagnosed glioblastoma. J Neurooncol 2013;112:277–83.

8. Network TCGATR. Comprehensive genomic characterization defineshuman glioblastoma genes and core pathways. Nature 2008;455:1061–8.

9. https://clinicaltrials.gov.10. Rodon J, Dienstmann R, Serra V, Tabernero J. Development of PI3K

inhibitors: Lessons learned from early clinical trials. Nat Rev Clin Oncol2013;10:143–53.

B

3.04.0 ppm1.52.02.53.5

Cho

PC

GPC

Lac

AlaGlu,Gln Gly Tau

m-Ins

Cr, PCrm-InsLac m-Ins

A

Cho

PCGPC

Cho

GPCPC

Cho

GPCPC

Cho

GPCPC

0

0.1

0.2

0.3

0.4

0.5

0.6

CTRL

Voxt

alisib

Voxt

alisib

/TMZ TM

Z

tCho

(AU)

GPC PC Cho

Gln

Glx GluNAA

CTRL Voxtalisib Voxtalisib/TMZ

TMZ

Figure 6.U87-MG tumor model: A, representative ex vivo 1H HR-MAS spectra from a biopsy of control tumor. Insets exhibit spectra zoomed in on the choline region. B, tCholevels and individual choline-containing metabolites. Error bars, SD of the tCho.






11. Ma DJ, Galanis E, Anderson SK, Schiff D, Kaufmann TJ, Peller PJ, et al. Aphase II trial of everolimus, temozolomide, and radiotherapy in patientswith newly diagnosed glioblastoma: NCCTG N057K. Neuro Oncol2015;17:1261–9.

12. LassenU, SorensenM,Gaziel TB,Hasselbalch B, PoulsenHS. Phase II studyof bevacizumab and temsirolimus combination therapy for recurrentglioblastoma multiforme. Anticancer Res 2013;33:1657–60.

13. Kuger S, Graus D, Brendtke R, G€unther N, Katzer A, Lutyj P, et al. Radio-sensitization of glioblastoma cell lines by the dual PI3K and mTORinhibitor NVP-BEZ235 depends on drug-irradiation schedule. TranslOncol 2013;6:169–79.

14. Fan QW, Knight ZA, Goldenberg DD, Yu W, Mostov KE, Stokoe D, et al. Adual PI3 kinase/mTOR inhibitor reveals emergent efficacy in glioma.Cancer Cell 2006;9:341–9.

15. YuP, LairdAD,DuX,Wu J,WonKA,YamaguchiK, et al. Characterizationofthe activity of the PI3K/mTOR inhibitor XL765 (SAR245409) in tumormodels with diverse genetic alterations affecting the PI3K pathway. MolCancer Ther 2014;13:1078–91.

16. Papadopoulos KP, Tabernero J, Markman B, Patnaik A, Tolcher AW,Baselga J, et al. Phase I safety, pharmacokinetic, and pharmacodynamicstudy of SAR245409 (XL765), a novel, orally administered PI3K/mTORinhibitor in patients with advanced solid tumors. Clin Cancer Res2014;20:2445–56.

17. Salphati L, Heffron TP, Alicke B, Nishimura M, Barck K, Carano RA, et al.Targeting the PI3K pathway in the brain–efficacy of a PI3K inhibitoroptimized to cross the blood-brain barrier. Clin Cancer Res 2012;18:6239–48.

18. Wan X, Harkavy B, Shen N, Grohar P, Helman LJ. Rapamycin inducesfeedback activation of Akt signaling through an IGF-1R-dependent mech-anism. Oncogene 2007;26:1932–40.

19. Figlin RA, Kaufmann I, Brechbiel J. Targeting PI3K and mTORC2 inmetastatic renal cell carcinoma: New strategies for overcoming resistanceto VEGFR and mTORC1 inhibitors. Int J Cancer 2013;133:788–96.

20. Chinot OL, Macdonald DR, Abrey LE, Zahlmann G, Kerlo€eguen Y,Cloughesy TF. Response assessment criteria for glioblastoma: practicaladaptation and implementation in clinical trials of antiangiogenic therapy.Curr Neurol Neurosci Rep 2013;13:347.

21. Courtney KD, Corcoran RB, Engelman JA. The PI3K pathway as drug targetin human cancer. J Clin Oncol 2010;28:1075–83.

22. Maj�osa C, Juli�a-Sap�eb M, Alonsoa J, Serrallongaa M, Aguileraa C,Acebesc JJ, et al. Brain tumor classification by proton MR spectroscopy:Comparison of diagnostic accuracy at short and long TE. Am J Neuror-adiol 2004;25:1696–704.

23. Zhang H, Ma L, Wang Q, Zheng X, Wu C, Xu BN. Role of magneticresonance spectroscopy for the differentiation of recurrent glioma fromradiation necrosis: a systematic review and meta-analysis. Eur J Radiol2014;83:2181–9.

24. Nelson SJ. Multivoxel magnetic resonance spectroscopy of brain tumors.Mol Cancer Ther 2003;2:497.

25. Glunde K, Bhujwalla ZM, Ronen SM. Choline metabolism in malignanttransformation. Nat Rev Cancer 2011;11:835–48.

26. VenkateshHS,ChaumeilMM,WardCS,Haas-KoganDA, JamesCD,RonenSM. Reduced phosphocholine and hyperpolarized lactate provide mag-netic resonance biomarkers of PI3K/Akt/mTOR inhibition in glioblasto-ma. Neuro Oncol 2012;14:315–25.

27. Beloueche-Babari M, Jackson LE, Al-Saffar NM, Eccles SA, Raynaud FI,Workman P, et al. Identification of magnetic resonance detectable meta-bolic changes associated with inhibition of phosphoinositide 3-kinasesignaling in human breast cancer cells. Mol Cancer Ther 2006;5:187–96.

28. Koul D, Shen R, Kim YW, Kondo Y, Lu Y, Bankson J, et al. Cellular andin vivo activity of a novel PI3K inhibitor, PX-866, against human glio-blastoma. Neuro Oncol 2010;12:559–69.

29. Ardenkjaer-Larsen JH, Fridlund B, Gram A, Hansson G, Hansson L, LercheMH, et al. Increase in signal-to-noise ratio of >10,000 times in liquid-stateNMR. Proc Natl Acad Sci U S A 2003;100:10158–63.

30. Kurhanewicz J, Vigneron DB, Brindle K, Chekmenev EY, Comment A,Cunningham CH, et al. Analysis of cancer metabolism by imaging hyper-polarized nuclei: Prospects for translation to clinical research. Neoplasia2011;13:81–97.

31. Brindle K. Watching tumours gasp and die with MRI: The promise ofhyperpolarised 13C MR spectroscopic imaging. Br J Radiol 2012;85:697–708.

32. Chaumeil MM, Ozawa T, Park I, Scott K, James CD, Nelson SJ, et al.Hyperpolarized 13C MR spectroscopic imaging can be used to monitoreverolimus treatment in vivo in an orthotopic rodent model of glioblas-toma. Neuroimage 2012;59:193–201.

33. Ward CS, Venkatesh HS, Chaumeil MM, Brandes AH, Vancriekinge M,Dafni H, et al. Noninvasive detection of target modulation followingphosphatidylinositol 3-kinase inhibition using hyperpolarized 13C mag-netic resonance spectroscopy. Cancer Res 2010;70:1296–305.

34. Park I, Bok R, Ozawa T, Phillips JJ, James CD, Vigneron DB, et al. Detectionof early response to temozolomide treatment in brain tumors usinghyperpolarized 13C MR metabolic imaging. J Magn Reson Imaging 2011;33:1284–90.

35. Park I, Mukherjee J, Ito M, Chaumeil MM, Jalbert LE, Gaensler K, et al.Changes in pyruvate metabolism detected by magnetic resonance imagingare linked toDNAdamage and serve as a sensor of temozolomide responsein glioblastoma cells. Cancer Res 2014;74:7115–24.

36. Nelson SJ, Kurhanewicz J, Vigneron DB, Larson PEZ, Harzstark AJ, FerroneM, et al. Metabolic imaging of patients with prostate cancer using hyper-polarized [1-13C]pyruvate. Sci Transl Med 2013;5:198.

37. Nelson SJ. Analysis of volumeMRI andMR spectroscopic imaging data forthe evaluation of patients with brain tumors. Magn Reson Med 2001;46:228–39.

38. Crane JC, Olson MP, Nelson SJ. SIVIC: Open-source, standards-basedsoftware for DICOM MR spectroscopy workflows. Int J Biomed Imaging2013;2013:169526.

39. jMRUI. http://www.mrui.uab.es/mrui/mrui_Overview.shtml.40. Mnova. http://mestrelab.com/.41. Dafni H, Larson PE, Hu S, Yoshihara HA, Ward CS, Venkatesh HS, et al.

Hyperpolarized 13C spectroscopic imaging informs on hypoxia-induciblefactor-1 and myc activity downstream of platelet-derived growth factorreceptor. Cancer Res 2010;70:7400–10.

42. MasonWP,Macneil M, Kavan P, Easaw J, Macdonald D, Thiessen B, et al. Aphase I study of temozolomide and everolimus (RAD001) in patients withnewly diagnosed and progressive glioblastoma either receiving or notreceiving enzyme-inducing anticonvulsants: an NCIC CTG study. InvestNew Drugs 2012;30:2344–51.

43. Prasad G, Sottero T, Yang X, Mueller S, James CD, Weiss WA, et al.Inhibition of PI3K/mTOR pathways in glioblastoma and implicationsfor combination therapy with temozolomide. Neuro Oncol 2011;13:384–92.

44. Esmaeili M, Bathen TF, Engebra�ten O, Mælandsmo GM, Gribbestad IS,

Moestue SA.Quantitative (31)PHR-MASMR spectroscopy for detection ofresponse to PI3K/mTOR inhibition in breast cancer xenografts. MagnReson Med 2014;71:1973–81.

45. Al-Saffar NM, Marshall LV, Jackson LE, Balarajah G, Eykyn TR, Agliano A,et al. Lactate and cholinemetabolites detected in vitro by nuclear magneticresonance spectroscopy are potential metabolic biomarkers for PI3Kinhibition in pediatric glioblastoma. PLoS One 2014;9:e103835.

46. Al-Saffar NM, Jackson LE, Raynaud FI, Clarke PA, Ramírez de Molina A,Lacal JC, et al. The phosphoinositide 3-kinase inhibitor PI-103 down-regulates choline kinase alpha leading to phosphocholine and total cho-line decrease detected by magnetic resonance spectroscopy. Cancer Res2010;70:5507–17.

47. Sandulache VC, Chen Y, Skinner HD, Lu T, Feng L, Court LE, et al. Acutetumor lactate perturbations as a biomarker of genotoxic stress: develop-ment of a biochemical model. Mol Cancer Ther 2015;14:2901–8.

48. LinG, AndrejevaG,Wong Te FongAC,Hill DK,OrtonMR, ParkesHG, et al.ReducedWarburg effect in cancer cells undergoing autophagy: Steady-state1H-MRS and real-time hyperpolarized 13C-MRS studies. PLoS One2014;9:e92645.

49. Lodi A, Woods SM, Ronen SM. Treatment with the MEK inhibitor U0126induces decreased hyperpolarized pyruvate to lactate conversion in breastbut not in prostate cancer cells. NMR Biomed 2013;26:299–306.

50. Wu SH, Bi JF, Cloughesy T, CaveneeWK,Mischel PS. Emerging function ofmTORC2 as a core regulator in glioblastoma: metabolic reprogrammingand drug resistance. Cancer Biol Med 2014;11:255–63.


Radoul et al.




2016;15:1113-1122. Published OnlineFirst February 16, 2016.Mol Cancer Ther Marina Radoul, Myriam M. Chaumeil, Pia Eriksson, et al. with Enhanced SurvivalInhibitor and Temozolomide:Metabolic Changes Are Associated MR Studies of Glioblastoma Models Treated with Dual PI3K/mTOR

Updated version

10.1158/1535-7163.MCT-15-0769doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2016/02/13/1535-7163.MCT-15-0769.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/15/5/1113.full#ref-list-1

This article cites 46 articles, 15 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/15/5/1113.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/15/5/1113To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-15-0769

http://mct.aacrjournals.org/content/suppl/2016/02/13/1535-7163.MCT-15-0769.DC1

http://mct.aacrjournals.org/content/15/5/1113.full#ref-list-1

http://mct.aacrjournals.org/content/15/5/1113.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/15/5/1113


MR Studies of Glioblastoma Models Treated with Dual … · Tumor Registry of the United States for...

Documents

Transcript of MR Studies of Glioblastoma Models Treated with Dual … · Tumor Registry of the United States for...