Bevacizumab-Induced Inhibition of Angiogenesis Promotes a ... · expressed as best tumor growth inhibition [%T/C ¼ (median volume of treated tumors/median volume of control tumors)

Cancer Biology and Signal Transduction

Bevacizumab-Induced Inhibition of AngiogenesisPromotes a More Homogeneous IntratumoralDistribution of Paclitaxel, Improving theAntitumor ResponseMarta Cesca1, Lavinia Morosi1, Alexander Berndt2, Ilaria Fuso Nerini1, Roberta Frapolli1,Petra Richter2, Alessandra Decio1, Olaf Dirsch2, Edoardo Micotti3, Silvia Giordano4,Maurizio D'Incalci1, Enrico Davoli4, Massimo Zucchetti1, and Raffaella Giavazzi1

Abstract

The antitumor activity of angiogenesis inhibitors is rein-forced in combination with chemotherapy. It is debated wheth-er this potentiation is related to a better drug delivery to thetumor due to the antiangiogenic effects on tumor vesselphenotype and functionality. We addressed this question bycombining bevacizumab with paclitaxel on A2780-1A9 ovariancarcinoma and HT-29 colon carcinoma transplanted ectopical-ly in the subcutis of nude mice and on A2780-1A9 and IGROV1ovarian carcinoma transplanted orthotopically in the bursa ofthe mouse ovary. Paclitaxel concentrations together with itsdistribution by MALDI mass spectrometry imaging (MALDIMSI) were measured to determine the drug in different areasof the tumor, which was immunostained to depict vesselmorphology and tumor proliferation. Bevacizumab modifiedthe vessel bed, assessed by CD31 staining and dynamic contrast

enhanced MRI (DCE-MRI), and potentiated the antitumoractivity of paclitaxel in all the models. Although tumor paclitaxelconcentrations were lower after bevacizumab, the drug distribut-ed more homogeneously, particularly in vascularized, non-necrotic areas, and was cleared more slowly than controls. Thishappened specifically in tumor tissue, as there was no change inpaclitaxel pharmacokinetics or drug distribution in normal tis-sues. In addition, the drug concentration and distribution werenot influenced by the site of tumor growth, as A2780-1A9 andIGROV1 growing in the ovary gave results similar to the tumorgrowing subcutaneously.We suggest that the changes in the tumormicroenvironment architecture induced by bevacizumab, togeth-er with the better distribution of paclitaxel, may explain thesignificant antitumor potentiation by the combination.Mol CancerTher; 15(1); 125–35. �2015 AACR.

IntroductionAngiogenesis-targeting agents are currently used in cancer

treatment (1, 2). Targeting the VEGF–VEGFR signaling axis withligand-binding drugs (i.e., with the antibody bevacizumab) orwith low-molecular-weight VEGF receptor tyrosin kinase inhibi-tors (TKI) shows efficacy in different tumor types, though thesurvival benefit associated to these class of drugs is modest. Atvariance with the TKI, which are in general given as single-agenttherapy, bevacizumab, in most settings, requires the addition ofcytotoxic therapy (2, 3). It has particularly beneficial effect whencombined with chemotherapy, prolonging overall survival (OS)

in patients with metastatic colorectal cancer (4), recurrent/advanced non–small cell lung cancer (NSCLC; ref. 5), and—morerecently—with advanced cervical cancer (6). Bevacizumab, com-binedwith chemotherapy, and in amaintenance regimen, extend-ed progression-free survival in patients with advanced ovariancancer (7, 8).

The mechanism by which antiangiogenic agents can boost theefficacy of chemotherapy is not completely understood, as thesedrugs can act on different tumor compartments, not necessarilylinked (9–12). Several hypotheses have been proposed. Antian-giogenic drugs mainly have an antivascular effect, so that abnor-mal, inefficient vessels are destroyed and the remainder areremodeled toward a more "mature" or "normal" phenotype(13), which in turn can lead to better delivery of therapeutics.Additional mechanisms are related to their action as chemosen-sitizing agents by inhibiting chemotherapy-inducedmobilizationand tumor "homing" of proangiogenic bone marrow–derivedcirculating endothelial progenitors (14), or to chemotherapy-enhanced damage to endothelial cells when prosurvival signalsof VEGF are blocked (15, 16). We showed induction of tumornecrosis by a vascular distrupting agent and increased prolifera-tion in the remaining viable tumor tissue where chemotherapymay be more active (17).

In terms of drug delivery, giving chemotherapy after an anti-angiogenic agent should not sound rational, since when thetumor vasculature is modified or even disrupted the delivery of

1Department of Oncology, IRCCS-Istituto di Ricerche FarmacologicheMario Negri, Milan, Italy. 2Institute of Pathology, University Hospital,Jena, Germany. 3Department of Neuroscience, IRCCS-Istituto diRicerche Farmacologiche Mario Negri, Milan, Italy. 4Department ofEnvironmental Health, IRCCS-Istituto di Ricerche FarmacologicheMario Negri, Milan, Italy.

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

Corresponding Author: Raffaella Giavazzi, IRCCS—Istituto di Ricerche Farm-acologiche Mario Negri, Via Giuseppe La Masa 19, 20156 Milan, Italy. Phone: 39-02-3901-4732; Fax: 39-02-3901-4734; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-15-0063

�2015 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org 125

on July 26, 2020. © 2016 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst October 22, 2015; DOI: 10.1158/1535-7163.MCT-15-0063

http://mct.aacrjournals.org/

cytotoxic drugs may be impaired and the response to therapycannot improve. Normalization of the tumor vasculature (18, 19)offers a solution to this apparent paradox. An appropriate dose ofantiangiogenic agents would transiently normalize the tumorvasculature, leading to better blood flow and lower interstitialfluid pressure (IFP), permitting an increase in drug delivery, thusimproving the outcome of the therapy (13, 19, 20). Numerouspreclinical studies, including ours, have shown the effect ofantiangiogenic therapy on vessel morphology and function.However, experimental evidence of an enhanced antitumor effectof the combination due to better tumor drug uptake is scant anddebated (21–23). There are few clinical studies reporting the effectof antiangiogenic therapy on tumor drug concentrations. Beva-cizumabhad antivascular and normalizing action in patientswithrectal carcinoma, indirectly indicating that blood vessels weremore efficient after bevacizumab (24). On the other hand, it hasalso been reported that bevacizumab did not improve tumor drugdelivery, and in fact had the opposite effect (25).

The scarcity of data on this issue prompted this study in whichwe investigated the relationship between the pharmacologiceffects of bevacizumab and the tumor penetration of paclitaxel,combined. Using conventional analytic techniques based onliquid chromatography (LC) or LC coupled to mass spectrometry(LC/MS),which gives informationon the total drug concentrationinside the tumor, together with matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MALDI MSI),which indicates the drug distribution in various compartments ofthe tissue (26, 27), we examined the intratumor distribution ofpaclitaxel, which is probably a key factor in the antitumor effect ofthe combination.

Our findings indicate that the greater antitumor activity ofpaclitaxel after bevacizumab is not necessarily due to overallincreaseddrugdelivery, but to the restorationof amore functionaltumormicroenvironment architecture that facilitates distributionof the cytotoxic drug in more vital, actively proliferating areas.

Materials and MethodsDrug preparation and administration

Bevacizumab(Avastin; RocheS.p.A.)wasdiluted in saline beforeuse, and injected i.p. at the dose of 150 mg/mouse twice a week. Forantitumor activity paclitaxel (Indena S.p.A.) was dissolved in50% Cremophor EL (Sigma) and 50% ethanol, further dilutedwith saline immediately before use, and injected i.v. at a dose of20 mg/kg weekly for three courses (Q7�3), or 60 mg/kg (singlebolus), as specified in the Results. Doxorubicin (NervianoMedicalScience) was dissolved in water and injected i.v. at the dose of6 mg/kg Q7�3.

For pharmacokinetics and MALDI MSI, mice were given singlebolus injections of chemotherapy, as detailed below. Controlmice received the vehicles.

AnimalsSix- to 8-week-old female NCr-nu/numice were obtained from

Harlan Laboratories. They were maintained under specific path-ogen-free conditions, housed in isolated vented cages, and han-dled using aseptic procedures. Procedures involving animals andtheir care were conducted in conformity with institutional guide-lines that comply with national (Legislative Decree 26, March,2014) and international (EECCouncilDirective 2010/63, August,2013) laws andpolicies, in linewith guidelines for thewelfare and

use of animals in cancer research (28). Animal studies wereapproved by the Mario Negri Institute Animal Care and UseCommittee and by Italian Ministerial decree no. 84-2013.

Tumor cell linesTheA2780-derived-1A9(A2780-1A9)humanovarian cancer cell

line (29)was kindly provided by Tito Fojo in 1994 (NCI-NIH) andthe IGROV1humanovarian cancer cell line (30)wasobtained fromthe National Cancer Institute Tumor Repository in 1995. 1A9-lucand IGROV1-luc are their variants infected with lentiviral vectorcarrying the coding sequence of syntheticfirefly luciferase gene, luc2(Photinus pyralis; refs. 31, 32). The HT-29 colon cancer line wasobtained from the ATCC in the 1980s (33). Authentication of thecell lines was performed by comparing their DNA short tandemrepeat profile (AmpFlSTR identifiler Plus PCR Amplification kit,Applied Biosystems) with a DNA fingerprinting database. Cellcultures were maintained in RPMI-1640 (BioWest) supplementedwith 10% FBS (Euroclone). Master stocks of the cell lines werestored frozen in liquidnitrogen, newampoules thawed at need andcultures maintained for no more than 6 weeks before use.

Xenograft tumor models and antitumor activityEctopic models. A2780-1A9 human ovarian carcinoma (10� 106

cells), or HT-29 human colon carcinoma (4 � 106 cells), wereimplanted s.c. in the flank of nude mice (34, 35).

Tumor growth was measured with a digital caliper two/threetimes a week and tumor volume (mm3) was calculated as [length(mm) � width2 (mm2)]/2. We used the Study Director 2.1software (Studylog System, Inc.) for animal management anddata collection.

Results were plotted as relative tumor volume (RTV), calculatedwith the formula Vt/V0, where Vt is the tumor volume on any dayof measurement and V0 is the tumor volume at the beginning oftreatment. For A2780-1A9 the efficacy of the treatment wasexpressed as best tumor growth inhibition [%T/C ¼ (medianvolume of treated tumors/median volume of control tumors) �100], or tumor log cell kill (LCK), with the formula GD/(3.32 �DT), where GD (growth delay) is the difference in days requiredfor the treated tumors to reach a target size (1,500 mm3) com-pared with the control group, and DT is the tumor doubling time.The combinations were considered additive, synergistic or antag-onist if the LCK (LCKob) was, respectively, equal to, larger, orsmaller than the expected LCK (LCKexp), calculated as the sum ofthe LCK of each single therapy. Animals were euthanized whenmean primary tumor volume was �2,000 mm3.

Orthotopic models.A2780-1A9 (1A9-luc variant) and IGROV1-luchuman ovarian carcinoma cells (1 � 106) expressing the fireflyluciferase gene luc-2 were injected orthotopically under the bursaof the mouse ovary, as previously reported (31, 32). Biolumines-cence imaging (BLI) was used to confirm the presence of tumor inthe ovary, to randomizemice at the beginning of treatment and tofollow tumor progression as described in refs. (31, 32). Briefly, D-luciferin (150 mg/kg, i.p., Caliper Lifescience)–injected micewere scanned after 10 minutes with eXplore Optix MX2 (ART,Advanced Research Technologies Inc.) and images analyzed withOptiview software (ART, Advanced Research Technologies Inc.).Photon counts were indicative of tumor burden. For antitumoractivity mice were killed at the first signs of discomfort (the dayof death being considered the limit of survival) and resultsplotted as the percentage survival against days after tumor

Cesca et al.

Mol Cancer Ther; 15(1) January 2016 Molecular Cancer Therapeutics126




transplant. The increment of lifespan (ILS) was calculated as 100� [(median survival dayof treated group�median survival day ofcontrol group)/median survival day of control group].

Pharmacokinetic studiesTumor-bearing mice (4/5 per time point) were given paclitaxel

(20 or 60mg/kg, day 0) or doxorubicin (6mg/kg, day 0), alone orafter bevacizumab (two injections, days �5 and �1). Biologicspecimens were collected 0.25, 2, 6, 8, 24, 48, 72 hours after 20mg/kg paclitaxel or 6 hours after paclitaxel in the 60mg/kg study,or 0.08, 0.25, 0.5, 1, 3, 6, 24, 48, 72hours after doxorubicin. Blood

was taken from the retro-orbital plexus under isoflorane anesthe-sia and collected in heparinized tubes. Animals were killed andtumors and liver were removed and immediately frozen at�80�Cuntil analysis of the drug concentration. The plasma fraction wasimmediately separated by centrifugation (4,000 rpm, 15minutes,4�C) and stored at �20�C until analysis. Paclitaxel levels weredeterminedbyhigh-performance liquid chromatography (HPLC)with UV detection, modified for measurements in plasma andtissues (34). The total concentration of doxorubicin was deter-mined by HPLC coupled to fluorescence detection (lexc 475 nm;lem 550 nm) according to ref. 36.

Figure 1.Antitumor activity and tumor uptake of chemotherapy in combination with bevacizumab in A2780-1A9 subcutaneous tumor-bearing mice. A, mice receivedbevacizumab (Bev, ~, 150 mg/mouse, i.p., twice a week, from day �5, median tumor volume 150 mm3), alone or in combination with paclitaxel. Mice wererandomized to start paclitaxel (PTX, ~, 20 mg/kg, i.v) 24 hours after the second injection of bevacizumab (day 0) with a median tumor volume of 400 mm3.This schedule was repeated for 3 weeks, giving chemotherapy on days 0, 7, and 14 (Q7 � 3). Values are RTV (mean � SEM, 10/group; ��, P < 0.0001 vs.chemotherapy alone). B, paclitaxel concentration (PTX, 20 mg/kg i.v., day 0) in tumors (left), livers and plasma (right), at 0.25, 2, 6, 8, 24, 48, and 72 hours,in mice pretreated with bevacizumab (Bev, 150 mg/mouse) or vehicle on days �5 and �1 (mean � SEM, 4/group; �� , P < 0.01; �� , P < 0.0001). C, mice receivedbevacizumab (Bev, ~, 150 mg/mouse, i.p., twice a week, from day �5, median tumor volume 150 mm3), alone or in combination with doxorubicin. Mice wererandomized to start doxorubicin (Doxo, ~, 6 mg/kg, i.v.) 24 hours after the second injection of bevacizumab (day 0) with a median tumor volume of 400 mm3.This schedule was repeated for 3 weeks, giving chemotherapy on days 0, 7, and 14 (Q7 � 3). Values are RTV (mean � SEM, 10/group; ��, P < 0.0001, vs.chemotherapy alone). D, doxorubicin concentration (Doxo, 6mg/kg i.v., day0) in tumors (left), livers andplasma (right), at 0.08, 0.25, 0.5, 1, 3, 6, 24, 48, and 72 hours,in mice pretreated with bevacizumab or vehicle on days �5 and �1 (mean � SEM, 4/group; � , P < 0.05; �� , P < 0.001). E, DCE-MRI in tumor-bearing mice24 hours after vehicle or bevacizumab (Bev, 150 mg/mouse, i.p., days �5 and �1). Left, tumor IAUC 5 is the initial AUC in the tumor region calculated fromthe first 5minutes of theDCE-MRI signal time course (box plot, 6/group; �� , P <0.01). Right, representative images of the contrast agent enhancement superimposedover anatomical MRI images of tumors from a vehicle and a bevacizumab-treated mouse. F, vessel analysis from tumor harvested 24 hours after receivingvehicle or bevacizumab (Bev, 150 mg/mouse, i.p., days �5 and �1). From left to right, diameter, area, and number of vessels measured after immunostainingfor CD31 (mean � SEM, 5/group; �� , P < 0.001; �� , P < 0.0001); right, representative images of A2780-1A9 tumors stained for CD31 (scale bar, 200 mm).

Paclitaxel Tumor Distribution after Bevacizumab

www.aacrjournals.org Mol Cancer Ther; 15(1) January 2016 127




Pharmacokinetic parameters were calculated using PhoenixWinNonlin 6.3 software (Pharsight).

MALDI mass spectrometry imaging analysisMice bearing different tumormodels (12/group), pretreated or

not with bevacizumab (two injections, days �5 and �1) weregiven 60mg/kg of paclitaxel (day 0). After 6 hours, that is close tothe tumor Cmax, tumors from five mice per group were collected,divided into two, and frozen in liquid nitrogen. The remaining 7mice were followed for antitumor activity. For imaging analysis,half frozen tumor tissue was cut into 10-mm-thick sections(three from each tumor) at�20�C using a cryo-microtome (LeicaMicrosystems) and mounted on pre-cooled MALDI plates (Opti-TOF 384Well insert) by standard thaw-mounting techniques. Foreach section, three adjacent slices (7-mm-thick) were cut forimmunohistochemical analysis. Paclitaxel MALDI imaging anal-ysis was done with a MALDI 4800 TOF-TOF (AB SCIEX OldConnecticut Path), as described (27), with details in Supplemen-tary Methods. The distribution of the paclitaxel ion signal in thenormalized images was analyzed by the software ImageJ (imagej.nih.gov/ij) as specified in Supplementary Methods.

The other half of the frozen tumor tissue was analyzed for totalpaclitaxel content by HPLC, as described above.

ImmunohistochemistryTo assess vessel density and size (CD31, MEC13.3 antibody,

Beckton Dickinson GmbH), tumor proliferation (Ki-67, SP6antibody, Thermo Fisher Scientific GmbH) and necrosis aftertreatments, tumors were excised, embedded in optimal cuttingcompound, snap-frozen, and stored at �80�C until immunohis-tochemical analysis (34). Tumor slices adjacent to MALDIimages were analyzed for CD31 and Ki-67 to overlay vesseldistribution/morphology and tumor cell proliferation/necrosiswith MALDI-imaged paclitaxel distribution. Detailed proceduresare described in SupplementaryMethods. For documentation andcomparison with MALDI MSI images, whole slides were scannedwith the Hamamatsu NonoZoomer 2.0HT and processed withNDP.view2 Software (Hamamatsu Photonics). Image analyseswere performed with computer-aided image analysis software(AxioVision Rel. 4.8.2; Zeiss).

In vivo magnetic resonance imagingDynamic contrast enhanced MRI (DCE-MRI) was done on

A2780-1A9 tumor-bearing mice using a BioSpec AVIII system(Bruker BioSpin; ref. 35). Tumor-bearing mice (6/group, vol-

ume matched), were pretreated or not with bevacizumab (twoinjections, days �5 and �1); 24 hours after the last dose theywere given gadopentetate dimeglumine Magnevist (Bayer,0.125 mmol/kg, i.v.) and scanned, as detailed in Supplemen-tary Methods. Image post-processing and analysis were done byin-house developed routines running under ImageJ (imagej.nih.gov/ij). IAUC was calculated as the initial area under thesignal enhancement curve in the manually selected tumorregion calculated from the first 5 minutes of the DCE-MRIsignal time course.

Statistical analysisStatistical analyses were done using Prism Software (GraphPad

Prism 6 Software). Differences in tumor growth and pharmaco-kinetic profiles were analyzed by two-way ANOVA followed bythe Tukey's post-test. The Student unpaired t test or the nonpara-metric Mann–Whitney test was used for the other cases. Differ-ences in survival were analyzed by the log-rank test. The P value<0.05 was considered significant.

Results and DiscussionBevacizumab impaired the tumor uptake of chemotherapydrugs, despite slower efflux, but increased the antitumoractivity of the combination

The main clinical use of bevacizumab in oncology is withchemotherapy (2, 3). Chemotherapy (including paclitaxel) com-bined with bevacizumab is becoming a standard of care forpatientswith ovarian cancer (37).We found that the combinationof bevacizumab with paclitaxel had significantly greater effect onA2780-1A9 ovarian tumor xenografts than the single drugs(Fig. 1A). In the paclitaxel and bevacizumab groups, there wasmoderate antitumor activity (T/C% 51, 62; LCK 4.2, 3.3, respec-tively), whereas the combination gave a T/C% of 40, indicating asynergistic effect (LCKexp 7.5, LCKob 14.2).

To assess whether the antitumor activity of the combinationwas due to delivery of paclitaxel in the tumor, we measured itsconcentration in tumor-bearing animals pretreated with bevaci-zumab. The pharmacokinetics of paclitaxel in tumor, plasma andliver was studied at different points after a cycle of bevacizumab.Paclitaxel levels in tumor tissue were generally lower, particularlyat early time points (30 minutes throughout 24 hours) in thebevacizumab-pretreated than in the vehicle-pretreated groups(Fig. 1B, left; Table 1). Bevacizumab pretreatment did not affectthe systemic pharmacokinetics of paclitaxel, and plasma and liver

Table 1. Concentrations of paclitaxel and doxorubicin at different times and AUC in A2780-1A9 tumor and liver from mice pretreated or not with bevacizumab

Tissue Treatment Cmax C48 h C72 h AUC0–72 h

(mg/g) (mg/g) (mg/g) (mg/g � h)

Tumora Paclitaxel 6.30 � 0.47 3.13 � 0.50 2.59 � 0.12 273.6Bevacizumab þ paclitaxel 4.90 � 0.24 2.70 � 0.29 2.90 � 0.21 205.2

Livera Paclitaxel 95.70 � 3.73 ND ND 309.1Bevacizumab þ paclitaxel 116.8 � 3.31 ND ND 347.0

Tumorb Doxorubicin 2.10 � 0.22 1.46 � 0.08 1.02 � 0.06 111.8Bevacizumab þ doxorubicin 1.60 � 0.10 1.34 � 0.06 1.24 � 0.06 98.0

Liverb Doxorubicin 15.90 � 0.86 1.50 � 0.12 0.64 � 0.01 290.2Bevacizumab þ doxorubicin 17.10 � 0.71 1.40 � 0.04 0.75 � 0.08 281.1

NOTE: n ¼ 4 mice/group (mean � SEM).Abbreviations: AUC, area under the curve of the concentration versus time point; Cmax, maximal drug concentration; C48h and C72h, drug concentrations at 48 and72 hours; ND, not detectable.aExperiment was run as in Fig. 1B.bExperiment was run as in Fig. 1D.

Cesca et al.





concentrations of paclitaxel were comparable in the two groups(Fig. 1B, right), thus not explaining the observations in tumortissue.

In this preclinical setting, a similar pharmacokinetic profile wasobtained with doxorubicin—another low-molecular-weight che-motherapy drug with different pharmacologic properties from

Figure 2.MALDI MSI analysis of paclitaxel in subcutaneous A2780-1A9 tumors from mice pretreated with bevacizumab. Mice bearing subcutaneous A2780-1A9 tumorswere randomized at a median tumor volume of 400 mm3 to receive a single dose of paclitaxel (PTX; 60 mg/kg, i.v., day 0) either alone or after bevacizumab(Bev, 150 mg/mouse, i.p., days �5 and �1). Tumors were excised 6 hours later. A and B, from top to bottom, overlapping whole-tumor MALDI MSI images ofpaclitaxel distribution, vessels (CD31), and proliferation/necrosis (Ki-67) in three adjacent slices of A2780-1A9 tumors, pretreatedwith vehicle (left), or bevacizumab(right). Representative images of three tumors/group. A, scale bar, 2.5 mm. B, enlargements as in A; scale bar, 500 mm.






paclitaxel—after bevacizumab. Doxorubicin levels in the A2780-1A9 tumor from mice receiving bevacizumab were lower than incontrol mice shortly after the injection, even though the combi-nation had greater antitumor activity (T/C% 51; LCKexp 5, LCKob

5.8 (Fig. 1C and D, left; Table 1). Again, liver and plasma levels ofdoxorubicin did not differ in the two groups (Fig. 1D, right). Weinfer that the lower uptake after bevacizumab is a common eventfor low-molecular-weight drugs.

A fewpreclinical studies have reported increased drug uptake inthe tumor after antiangiogenic treatments (22). Among others,improved tumor perfusion mirrored by increased drug deliveryafter bevacizumab-induced vascular remodeling was described inamodel of orthotopic neuroblastoma (38). In a colon carcinomamodel, the CPT11 concentration tended to be higher after VEGF-blocking therapy, paralleling increased tumor perfusion (39). Areduction in vascular permeability after bevacizumab withincreased paclitaxel concentration was described in a breast anda lung cancer xenograft models (40).

In our study though, drug concentrations were lower afterbevacizumab. Interestingly, the pretreatment slowed the drugefflux from tumors, as evidenced by the paclitaxel or doxorubicinconcentrations at 48 and 72 hours, which were comparable in thevehicle andbevacizumab-pretreated groups, in favor of antitumoractivity (Fig. 1B and D). This agrees with our previous findings ofreduced tumor uptake and slower efflux of paclitaxel, whencombinedwith different tyrosine kinase receptor (TKR) inhibitorsaffecting angiogenesis (vandetanib-targeting EGFR and VEGFR inan ovarian cancer model, sunitinib targeting VEGFR and PDGFRand the dual inhibitors of VEGFR2 and FGFR2, brivanib and

E-3810, in amammary cancer model; refs. 34, 41). Therefore, thiswas true for different tumor models and with different angiogen-esis inhibitors, where the outcome was always better with thecombinations.

Themostwidely known theory to explain the antitumor activityof the combination is that vessel changes after antiangiogenesistreatment (i.e., normalization of the tumor microenvironment)can ultimately improve drug delivery, explaining the betterresponse (13). Vessel normalization improves the supply ofblood and oxygenation into the tumor, and reduces the IFP, sothe delivery and penetration of therapeutics into tumor tissueimprove (19). In our A2780-1A9 tumor model, 24 hours afterbevacizumabDCE-MRI analysis showed a significant reduction intumor IAUC (Fig. 1E) 5 minutes after injection of the contrastagent. Thus, bevacizumab reduced the overall delivery of a con-trast agent to tumor tissue by altering tumor perfusion and/orpermeability, a finding in line with a reduced uptake of chemo-therapy (35). At this same time, CD31 staining of harvestedtumors showed significant reductions in vessel diameter and area,but not in density, indicating amore regular vasculature (Fig. 1F).Decreased tumor perfusion, with only reductions in vessel size/area has been reported in xenograft cancer models after acutetreatment with bevacizumab, whereas vessel density decreasedafter chronic treatment (42). The doses of the angiogenesisinhibitors—we used a low clinically relevant dose of bevacizu-mab—and duration of the treatment are likely to account for thedifferent results, in the different experimental settings(13, 42, 43). Inadequate vasculature and decreased permeabilitydue towrong timingor toohighdoseof an antiangiogenicsmay in

Figure 3.MALDI MSI quantification of paclitaxeldistribution, tumor morphologyassessment, paclitaxel tumor uptake, andantitumor activity. Mice bearingsubcutaneous A2780-1A9 tumors weretreated with paclitaxel (PTX) either aloneor after bevacizumab (Bev) as per protocolin Fig. 2; controls were vehicles andbevacizumab alone. Tumors were excised6 hours later (A–C and E), or mice werefollowed for antitumor activity (D). A, thepercentage of paclitaxel pixels above thethreshold calculated from MALDI MSIimages as shown in Fig. 2 (sum of twoindependent experiments, 8/group, mean� SEM; �� , P < 0.0001). B, the percentageof necrotic area (5/group, mean � SEM;�� , P < 0.01). C, paclitaxel concentration intumors measured by HPLC (sum of twoindependent experiments, 8/group, mean� SEM; �� , P <0.01). D, antitumor activity inmice receiving bevacizumab (Bev, ~),paclitaxel (PTX, ~), or the combination;RTV (7/group, mean � SEM; �� , P < 0.01;�� , P < 0.0001 vs. chemotherapy alone).E, the percentage of Ki-67–positiveproliferating tissue (5/group, mean� SEM;�� , P < 0.01; �� , P < 0.0001).

Cesca et al.





Figure 4.MALDI MSI analysis, paclitaxel tumor uptake, and antitumor activity in orthotopic IGROV1-luc ovarian cancer from mice pretreated with bevacizumab.Mice bearing IGROV1-luc tumors in the ovary were randomized by photon count (day 14 after transplant) to receive bevacizumab or not (Bev,150 mg/mouse, i.p., days 15 and 19), followed by paclitaxel (PTX, 60 mg/kg, i.v., day 20). Tumors were excised after 6 hours for MALDI MSI and histologicanalysis (A–E), or mice were followed for antitumor activity (F and G). A and B, from top to bottom, overlapping whole-tumor MALDI MSI images ofpaclitaxel distribution, vessels (CD31) and proliferation/necrosis (Ki-67) in three adjacent slices of IGROV1-luc tumors, pretreated with vehicle (left),or bevacizumab (right) and followed by paclitaxel 24 hours later; representative images of two tumors/group; scale bar, 2.5 mm. B, enlargements asin A, images representative of one tumor/group; scale bar, 500 mm. C, the percentage of paclitaxel pixels above the threshold calculated fromMALDI MSI images (5/group, mean � SEM; �� , P < 0.0001). D, vessel analysis (number and diameter) from MALDI MSI overlapping images (5/group,mean � SEM; �� , P < 0.001; �� , P < 0.0001). E, paclitaxel concentration in tumors measured by HPLC (5/group, mean � SEM). F, mice werechecked by BLI on day 18 (before receiving paclitaxel) and days 22, 35, and 42 (tumor progression). Left, the time course of tumor growth in onerepresentative mouse treated with paclitaxel (PTX) or bevacizumab plus paclitaxel (Bev þ PTX). Right, tumor burden is expressed as photoncounts (7/group, mean � SEM; �� , P < 0.01; �� , P < 0.001). G, survival of mice receiving paclitaxel (PTX), alone or after bevacizumab (Bev þ PTX; 7/group;�� , P < 0.01).






Figure 5.MALDI MSI analysis of paclitaxel, and antitumor activity in HT-29 tumors from mice pretreated with bevacizumab. Mice bearing subcutaneous HT-29colon carcinoma were randomized at a median tumor volume of 400 mm3 to receive a single dose of paclitaxel (PTX, 60 mg/kg, i.v., day 0) eitheralone or after bevacizumab (Bev, 150 mg/mouse, i.p., days �5 and �1). A, whole-tumor MALDI MSI images of paclitaxel distribution of HT-29 tumors,in mice pretreated with vehicle or bevacizumab and followed by paclitaxel 24 hours later. Tumors were excised 6 hours later; representativeimages of three tumors/group. (Continued on the following page.)

Cesca et al.





fact negatively affect the uptake of chemotherapy drug and ulti-mately the outcome of the combination therapy (13). Differenttumor models may also account for this discrepancy. We foundthat at the same experimental conditions bevacizumab signifi-cantly decreased vessel density in the IGROV1-luc ovarian and inthe HT-29 colon cancer models (see results in the paragraphbelow).

In line with our preclinical results, a human study in NSCLC,using positron emission tomography (PET), reported a rapid,significant reduction in perfusion and [11C]docetaxel uptake afterbevacizumab (25) and highlighted the importance of optimizingthe scheduling of antiangiogenic drugs. Reduced vascular perme-ability and flow by DCE-MRI was described after bevacizumab inpatients with advanced breast cancer (44).

These findings indicate that pretreatment with an antiangio-genic agent can lower the concentration of the anticancer drugused in combination. Nevertheless, here, the slower clearance—very likely due to vascular changes by bevacizumab—contributesto efficient tumor drug exposure.

MALDI MSI visualization of paclitaxel shows morehomogeneous intratumor distribution after bevacizumab

The reduced tumor drug concentrations after bevacizumab is inapparent contrast with the better antitumor effect of the combi-nation. This prompted us to investigate the spatial distribution ofpaclitaxel in tumors after bevacizumab, using a recentMALDIMSItechnology (27, 45) with which drug localization in differentareas of the tumor can be investigated. Visual inspection of theMALDI MSI images on A2780-1A9 tumor tissue sections frommice pretreated with bevacizumab showed more homogeneouspaclitaxel distribution than inmice pretreatedwith vehicle, wherethe drug distribution was irregular (Fig. 2A and B, top). There wasalso a significantly larger percentage of pixels above the back-ground noise threshold in sections from mice pretreated withbevacizumab than those from mice pretreated with vehicle, con-firming the better paclitaxel tumor distribution (Fig. 3A). Analysisof clusters of positive pixels in MALDI MSI tumor images showedthat in the sections frommice pretreated with bevacizumab therewere fewer aggregates of positive pixels and their mean size wasbigger than in sections from mice pretreated with vehicle (Sup-plementary Fig. S1). Together with the different percentages ofpixels above the background threshold these results highlight thebetter distribution of paclitaxel after bevacizumab pretreatmentand particularly the presence of the drug in larger and morehomogeneous areas of the tumor.

To find out whether the different distribution was linked todifferences in tissue architecture andmorphology, next wemade acomparative histologic analysis on adjacent tumor sections fromA2780-1A9, analyzed with MALDI MSI. Figure 2A (enlargedin Fig. 2B) shows the paclitaxel distribution overlapping opticalimages of adjacent tumor slices stained for CD31 (vessels) orKi-67 (proliferation). Unlike with vehicle, bevacizumab-pre-treated tumors had fewer necrotic areas (quantified in Fig. 3B),with vessels uniformly distributed throughout the tumor. This

indicated that the tumor distribution of paclitaxel was associatedwith vasculature and proliferating, non-necrotic areas, with amore homogeneous pattern in the bevacizumab-treated tumors,where paclitaxel was regularly distributed, with no breaks in thetissue. This sheds some light on the apparent paradox of thereduced paclitaxel uptake after bevacizumab, providing evidencethat the restoration of a functional tumor microenvironmentarchitecture modifies the distribution and retention of the anti-cancer drug in the tumor.

Part of the same tumor analyzed with MALDI MSI was used toquantify paclitaxel content by HPLC, confirming that in thissetting too total paclitaxel uptake in the tumors at 6 hours waslower after bevacizumab (Fig. 3C). Because we used 60 mg/kgpaclitaxel for the MALDI MSI, in a parallel trial we tested theantitumor activity of this dose combined with bevacizumab. Thecombination was significantly more active than the single treat-ments (Fig. 3D), as illustratedby the significantly lower number ofKi-67–positive cells in tumor sections from mice receiving pac-litaxel alone or the combination (Fig. 3E).

The site of tumor growth and the surrounding tumor envi-ronment can affect the biology of the tumor, influencing thedelivery and activity of certain drugs (46). This prompted us toinvestigate paclitaxel distribution in orthotopically growingA2780-1A9. Again, the distribution of paclitaxel measured byMALDI MSI after bevacizumab in tumors growing under thebursa of the ovary was more homogeneous than in controls(Supplementary Fig. S2A and S2B) and was associated with alow paclitaxel concentration by HPLC analysis (SupplementaryFig. S2C). The comparable results on A2780-1A9 tumor grow-ing ectopically in the subcutis or orthotopically in the ovarywere consistent with our previous finding, showing that thehistologic features (including vasculature) do not differ in thetwo sites (47).

To test the importance of these findings in another ovariantumor model, mice bearing orthotopic IGROV1-luc ovariantumor were pretreated or not with bevacizumab, before receivingpaclitaxel, then randomized for MALDI MSI and HPLC analysisand treatment evaluation. MALDI MSI clearly showed a better,more uniform paclitaxel distribution in tumors from mice pre-treated with bevacizumab than the vehicle-pretreated animals(Fig. 4A, top, enlarged in Fig. 4B); this was corroborated by thehigher percentage of pixels above the threshold and the analysis oftheir cluster (Fig. 4C and Supplementary Fig. S3A).

The corresponding histologic images stained for CD31 and Ki-67 (Fig. 4A, bottom, enlarged in Fig. 4B) show solidmicronodulargrowth of vehicle-treated tumors. Tumors are well vascularizedwith a characteristic continuous organization of the CD31-pos-itive vessel structures encircling the tumor complexes. Bevacizu-mab-pretreated tumors show a more homogeneous growthpattern with a loss of "perinodular" vessel formation and areduction in vessel number and diameter (quantified in Fig.4D). Both groups show moderate proliferative activity with anaccentuation of Ki-67–positive cells at the periphery of themicro-nodules (Fig. 4A and B, bottom).

(Continued.) B, the percentage of paclitaxel pixels above the threshold was calculated from MALDI MSI images (5/group, mean � SEM; �� , P < 0.0001).C, representative images of vessels (CD31) and proliferation (Ki-67) of tumors from MALDI MSI overlapping images, pretreated with vehicle (left) orbevacizumab (right), and followed by paclitaxel 24 hours later; scale bar, 500 mm. Vessel number (D) and the percentage of Ki-67–positive proliferating tissue(E; 5/group, mean � SEM; �� , P < 0.01; �� , P < 0.0001). F, antitumor activity in mice receiving paclitaxel (PTX), bevacizumab (Bev), or the combination.RTV—tumor fold change from randomization—6 and 10 days after chemotherapy (single mice, 7/group; � , P < 0.05; �� , P < 0.001 vs. chemotherapy alone).






At this time point, the paclitaxel concentration quantified byHPLC did not differ in the two groups, despite the significanthistological changes (Fig. 4E). Bevacizumab combined with pac-litaxel significantly delayed the growthof IGROV1-luc in theovary(Fig. 4F), and increased survival (ILS 21%) compared with che-motherapy alone (Fig. 4G). These findings confirm the antiangio-genic effect of bevacizumab, though with a distinct pattern ofvascular response in this model, and its ability to boost paclitaxelantitumor activity.

To test the significance of these findings in another tumor typefor which bevacizumab is clinically used (4), we analyzed theMALDIMSI spatial distribution of paclitaxel after bevacizumab inthe HT-29 colon cancer xenografts (Fig. 5). Again, paclitaxel iondistribution on HT-29 tumor sections from mice pretreated withbevacizumab appeared more homogeneous than in tumors fromvehicle-pretreated mice, as confirmed by the significantly higherpercentage of pixels above the threshold, and cluster size (Fig. 5Aand B; Supplementary Fig. S3B). The corresponding histologicimages show a reduction in CD31-stained vessels and Ki-67–positive cells in bevacizumab-pretreated tumors (Fig. 5C, quan-tified in Fig. 5D and E), which anticipated the antitumor effect ofpaclitaxel combined with bevacizumab evident from 6 days aftertreatment (Fig. 5F).

In conclusion, in bevacizumab-pretreated mice the tumordistribution of paclitaxel, shown by MALDI MSI (27), was morehomogeneous than in mice pretreated with vehicle, with slowerdrug clearance from the tumor tissue. The discrete distributionof paclitaxel in tumors pretreated or not with bevacizumab isrelated to different morphologic features of the tissue, as wenoted in microscopy IHC-stained tissue images superimposedon the MALDI MSI images. Drug penetration was inadequate inpoorly vascularized or necrotic parts of the control tumors, andwith the antiangiogenic pretreatment the anticancer drug wasconveyed to proliferating areas of the tumor, leading to moreeven distribution. This was true for different tumor modelsindependently of their origin or their site of implantation, andthe different pattern of vascular response after antiangiogenictreatment.

These findings suggest that the better therapeutic activity of thecombination is due not only to the sum of two drugs withdifferent modes of action, but also to a more favorable distribu-tion of the chemotherapy drug in the tumor. The better tumordistribution of paclitaxel is in keeping with its good antitumoractivity when it is combined with bevacizumab.

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

Authors' ContributionsConception and design: M. Cesca, R. Frapolli, M. D'Incalci, M. Zucchetti,R. GiavazziDevelopment of methodology: L. Morosi, A. Berndt, I.F. Nerini, P. Richter,A. Decio, E. Micotti, S. Giordano, E. DavoliAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.):M. Cesca, L. Morosi, R. Frapolli, O. Dirsch, E. Micotti,S. Giordano, R. GiavazziAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis):M. Cesca, L. Morosi, A. Berndt, R. Frapolli, P. Richter,A. Decio, S. Giordano, M. Zucchetti, R. GiavazziWriting, review, and/or revision of the manuscript: M. Cesca, A. Berndt,O. Dirsch, M. D'Incalci, M. Zucchetti, R. GiavazziAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): I.F. Nerini, E. DavoliStudy supervision: R. Giavazzi, M. Cesca

AcknowledgmentsThe authors thank Katja Gr€un for assistance in histologic analysis and J.D.

Baggott for editing the article. I. Fuso Nerini and A. Decio are the recipients ofa fellowship from the Italian Foundation for Cancer Research (FIRC).

Grant SupportThis study was supported by grants from the Italian Association for Cancer

Research (IG14532 and 12182; to R. Giavazzi) and the Fondazione CARIPLO(no. 2011-0614; to M. Cesca).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received January 22, 2015; revised October 5, 2015; accepted October 9,2015; published OnlineFirst October 22, 2015.

References1. Carmeliet P, Jain RK. Molecular mechanisms and clinical applications of

angiogenesis. Nature 2011;473:298–307.2. Jayson GC, Hicklin DJ, Ellis LM. Antiangiogenic therapy—evolving

view based on clinical trial results. Nat Rev Clin Oncol 2012;9:297–303.3. Moserle L, Jimenez-Valerio G, Casanovas O. Antiangiogenic therapies:

going beyond their limits. Cancer Discov 2014;4:31–41.4. Hurwitz H, Fehrenbacher L, NovotnyW, Cartwright T, Hainsworth J, Heim

W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin formetastatic colorectal cancer. N Engl J Med 2004;350:2335–42.

5. Sandler A, Gray R, Perry MC, Brahmer J, Schiller JH, Dowlati A, et al.Paclitaxel-carboplatin alone or with bevacizumab for non–small cell lungcancer. N Engl J Med 2006;355:2542–50.

6. Tewari KS, Sill MW, Long HJ III, Penson RT, Huang H, Ramondetta LM,et al. Improved survival with bevacizumab in advanced cervical cancer.N Engl J Med 2014;370:734–43.

7. Burger RA, BradyMF, BookmanMA, Fleming GF, Monk BJ, Huang H, et al.Incorporation of bevacizumab in the primary treatment of ovarian cancer.N Engl J Med 2011;365:2473–83.

8. Perren TJ, Swart AM, Pfisterer J, Ledermann JA, Pujade-Lauraine E, Kris-tensen G, et al. A phase 3 trial of bevacizumab in ovarian cancer. N EnglJ Med 2011;365:2484–96.

9. Gasparini G, Longo R, Fanelli M, Teicher BA. Combination of antiangio-genic therapy with other anticancer therapies: results, challenges, and openquestions. J Clin Oncol 2005;23:1295–311.

10. Moschetta M, Cesca M, Pretto F, Giavazzi R. Angiogenesis inhibitors:implications for combination with conventional therapies. Curr PharmDes 2010;16:3921–31.

11. Jain RK. Antiangiogenesis strategies revisited: from starving tumors toalleviating hypoxia. Cancer Cell 2014;26:605–22.

12. Boere IA, Hamberg P, Sleijfer S. It takes two to tango: combinationsof conventional cytotoxics with compounds targeting the vascularendothelial growth factor-vascular endothelial growth factor receptorpathway in patients with solid malignancies. Cancer Sci 2010;101:7–15.

13. Jain RK. Normalizing tumor microenvironment to treat cancer: bench tobedside to biomarkers. J Clin Oncol 2013;31:2205–18.

14. Shaked Y,Henke E, Roodhart JM,Mancuso P, LangenbergMH,ColleoniM,et al. Rapid chemotherapy-induced acute endothelial progenitor cellmobilization: implications for antiangiogenic drugs as chemosensitizingagents. Cancer Cell 2008;14:263–73.

15. Browder T, Butterfield CE, Kraling BM, Shi B, Marshall B, O'Reilly MS,et al. Antiangiogenic scheduling of chemotherapy improves effi-cacy against experimental drug-resistant cancer. Cancer Res 2000;60:1878–86.

16. Klement G, Baruchel S, Rak J, Man S, Clark K, Hicklin DJ, et al. Continuouslow-dose therapy with vinblastine and VEGF receptor-2 antibody inducessustained tumor regression without overt toxicity. J Clin Invest 2000;105:R15–24.


Cesca et al.




17. Martinelli M, Bonezzi K, Riccardi E, Kuhn E, Frapolli R, Zucchetti M,et al. Sequence dependent antitumour efficacy of the vascular disruptingagent ZD6126 in combination with paclitaxel. Br J Cancer 2007;97:888–94.

18. Le Serve AW, Hellmann K. Metastases and the normalization of tumourblood vessels by ICRF 159: a new type of drug action. Br Med J 1972;1:597–601.

19. Goel S, Duda DG, Xu L, Munn LL, Boucher Y, Fukumura D, et al. Nor-malization of the vasculature for treatment of cancer and other diseases.Physiol Rev 2011;91:1071–121.

20. Jain RK. Normalization of tumor vasculature: an emerging concept inantiangiogenic therapy. Science 2005;307:58–62.

21. Ma J, Waxman DJ. Combination of antiangiogenesis with chemotherapyfor more effective cancer treatment. Mol Cancer Ther 2008;7:3670–84.

22. Cesca M, Bizzaro F, Zucchetti M, Giavazzi R. Tumor Delivery of chemo-therapy combined with inhibitors of angiogenesis and vascular targetingagents. Front Oncol 2013;3:259.

23. Marcucci F, Corti A. How to improve exposure of tumor cells to drugs:promoter drugs increase tumor uptake and penetration of effector drugs.Adv Drug Deliv Rev 2012;64:53–68.

24. Willett CG, Boucher Y, di Tomaso E, Duda DG, Munn LL, Tong RT, et al.Direct evidence that the VEGF-specific antibody bevacizumab has anti-vascular effects in human rectal cancer. Nat Med 2004;10:145–7.

25. Van der Veldt AA, Lubberink M, Bahce I, Walraven M, de Boer MP, GreuterHN, et al. Rapid decrease in delivery of chemotherapy to tumors after anti-VEGF therapy: implications for scheduling of anti-angiogenic drugs. Can-cer Cell 2012;21:82–91.

26. Prideaux B, Stoeckli M. Mass spectrometry imaging for drug distributionstudies. J Proteomics 2012;75:4999–5013.

27. Morosi L, Spinelli P, Zucchetti M, Pretto F, Carra A, D'Incalci M, et al.Determination of paclitaxel distribution in solid tumors by nano-particleassisted laser desorption ionizationmass spectrometry imaging. PLoSONE2013;8:e72532.

28. Workman P, Aboagye EO, Balkwill F, Balmain A, Bruder G, Chaplin DJ,et al. Guidelines for the welfare and use of animals in cancer research. Br JCancer 2010;102:1555–77.

29. Belotti D, Rieppi M, Nicoletti MI, Casazza AM, Fojo T, Taraboletti G, et al.Paclitaxel (Taxol(R)) inhibits motility of paclitaxel-resistant human ovar-ian carcinoma cells. Clin Cancer Res 1996;2:1725–30.

30. Benard J, Da Silva J, De Blois MC, Boyer P, Duvillard P, Chiric E, et al.Characterization of a human ovarian adenocarcinoma line, IGROV1, intissue culture and in nude mice. Cancer Res 1985;45:4970–9.

31. Oliva P, Decio A, Castiglioni V, Bassi A, Pesenti E, Cesca M, et al. Cisplatinplus paclitaxel and maintenance of bevacizumab on tumour progression,dissemination, and survival of ovarian carcinoma xenograft models. Br JCancer 2012;107:360–9.

32. Decio A, Taraboletti G, Patton V, Alzani R, Perego P, Fruscio R, et al.Vascular endothelial growth factor c promotes ovarian carcinoma progres-sion through paracrine and autocrine mechanisms. Am J Pathol 2014;184:1050–61.

33. Price JE, Daniels LM, Campbell DE, Giavazzi R. Organ distribution ofexperimentalmetastases of a human colorectal carcinoma injected in nudemice. Clin Exp Metastasis 1989;7:55–68.

34. Cesca M, Frapolli R, Berndt A, Scarlato V, Richter P, Kosmehl H, et al. Theeffects of vandetanib on paclitaxel tumor distribution and antitumoractivity in a xenograft model of human ovarian carcinoma. Neoplasia2009;11:1155–64.

35. Bello E, ColellaG, Scarlato V,Oliva P, Berndt A, ValbusaG, et al. E-3810 is apotent dual inhibitor of VEGFR and FGFR that exerts antitumor activity inmultiple preclinical models. Cancer Res 2011;71:1396–405.

36. Colombo T, Zucchetti M, D'Incalci M. Cyclosporin Amarkedly changes thedistribution of doxorubicin in mice and rats. J Pharmacol Exp Ther 1994;269:22–7.

37. Banerjee S, Kaye SB. Gynecological cancer: first-line bevacizumab forovarian cancer—new standard of care? Nat Rev Clin Oncol 2012;9:194–6.

38. Dickson PV, Hamner JB, Sims TL, Fraga CH, Ng CY, Rajasekeran S, et al.Bevacizumab-induced transient remodeling of the vasculature in neuro-blastoma xenografts results in improved delivery and efficacy of system-ically administered chemotherapy. Clin Cancer Res 2007;13:3942–50.

39. Wildiers H, Guetens G, De Boeck G, Verbeken E, Landuyt B, Landuyt W,et al. Effect of antivascular endothelial growth factor treatment on theintratumoral uptake of CPT-11. Br J Cancer 2003;88:1979–86.

40. YanagisawaM, YorozuK, KurasawaM,NakanoK, Furugaki K, Yamashita Y,et al. Bevacizumab improves the delivery and efficacy of paclitaxel. Anti-cancer Drugs 2010;21:687–94.

41. Bello E, Taraboletti G, Colella G, Zucchetti M, Forestieri D, Licandro SA,et al. The tyrosine kinase inhibitor E-3810 combined with paclitaxelinhibits the growth of advanced-stage triple-negative breast cancer xeno-grafts. Mol Cancer Ther 2013;12:131–40.

42. Rajendran R, Huang W, Tang AM, Liang JM, Choo S, Reese T, et al. Earlydetection of antiangiogenic treatment responses in amouse xenograft tumormodel using quantitative perfusion MRI. Cancer Med 2014;3:47–60.

43. Hillman GG, Singh-Gupta V, Zhang H, Al-Bashir AK, Katkuri Y, Li M, et al.Dynamic contrast-enhanced magnetic resonance imaging of vascularchanges induced by sunitinib in papillary renal cell carcinoma xenografttumors. Neoplasia 2009;11:910–20.

44. Wedam SB, Low JA, Yang SX, Chow CK, Choyke P, Danforth D, et al.Antiangiogenic and antitumor effects of bevacizumab in patients withinflammatory and locally advanced breast cancer. J Clin Oncol 2006;24:769–77.

45. Buck A, Halbritter S, Spath C, Feuchtinger A, Aichler M, Zitzelsberger H,et al. Distribution andquantificationof irinotecan and its activemetaboliteSN-38 in colon cancer murine model systems using MALDI MSI. AnalBioanal Chem 2015;407:2107–16.

46. Fidler IJ, Wilmanns C, Staroselsky A, Radinsky R, Dong Z, Fan D. Mod-ulation of tumor cell response to chemotherapy by the organ environment.Cancer Metastasis Rev 1994;13:209–22.

47. Ricci F, Bizzaro F, CescaM, Guffanti F, Ganzinelli M, Decio A, et al. Patient-derived ovarian tumor xenografts recapitulate human clinicopathologyand genetic alterations. Cancer Res 2014;74:6980–90.






2016;15:125-135. Published OnlineFirst October 22, 2015.Mol Cancer Ther Marta Cesca, Lavinia Morosi, Alexander Berndt, et al. Antitumor Response

theHomogeneous Intratumoral Distribution of Paclitaxel, Improving Bevacizumab-Induced Inhibition of Angiogenesis Promotes a More

Updated version

10.1158/1535-7163.MCT-15-0063doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2015/10/21/1535-7163.MCT-15-0063.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/15/1/125.full#ref-list-1

This article cites 47 articles, 15 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/15/1/125.full#related-urls

This article has been cited by 3 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/15/1/125To 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-15-0063

http://mct.aacrjournals.org/content/suppl/2015/10/21/1535-7163.MCT-15-0063.DC1

http://mct.aacrjournals.org/content/15/1/125.full#ref-list-1

http://mct.aacrjournals.org/content/15/1/125.full#related-urls

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

mailto:[email protected]

http://mct.aacrjournals.org/content/15/1/125


Documents

Bevacizumab-Induced Inhibition of Angiogenesis Promotes a ... · expressed as best tumor growth inhibition [%T/C ¼ (median volume of treated tumors/median volume of control tumors)