PreclinicalEfﬁcacyforAKTTargetinginClearCell Carcinoma of ... · Carcinoma of the Ovary Tomoyuki...

Signal Transduction

Preclinical Efficacy for AKT Targeting in Clear CellCarcinoma of the OvaryTomoyuki Sasano1, Seiji Mabuchi1, Hiromasa Kuroda1, Mahiru Kawano1, Yuri Matsumoto1,Ryoko Takahashi1, Takeshi Hisamatsu1, Kenjiro Sawada1, Kae Hashimoto1, Aki Isobe1,Joseph R. Testa2, and Tadashi Kimura1

Abstract

The aim of this study was to determine the role of AKT as atherapeutic target in ovarian clear cell carcinoma (CCC), anaggressive, chemoresistant histologic subtype of ovarian can-cer. AKT activation was assessed by immunohistochemistry(IHC) using human tissue microarrays of primary ovariancancers, composed of both CCC and serous adenocarcinoma(SAC). The growth-inhibitory effect of AKT-specific targetingby the small-molecule inhibitor, perifosine, was examinedusing ovarian CCC cell lines in vitro and in vivo. Finally, theactivity of perifosine was examined using in CCC-derivedtumors that had acquired resistance to anti-VEGF or che-motherapeutics such as bevacizumab or cisplatin, respectively.Interestingly, AKT was frequently activated both in early-stageand advanced-stage CCCs. Treatment of CCC cells with peri-fosine attenuated the activity of AKT–mTORC1 signaling,inhibited proliferation, and induced apoptosis. The effect of

perifosine was more profound under conditions of high AKTactivity compared with low AKT activity. Increased AKT activa-tion and enhanced sensitivity to perifosine were observed in thecontext of cisplatin-resistant CCC. Treatment with perifosineconcurrently with cisplatin significantly enhanced the antitu-mor effect of cisplatin. Moreover, perifosine showed significantantitumor activity in CCC-derived tumors that had acquiredresistance to bevacizumab or cisplatin. Collectively, these datareveal that AKT is frequently activated in ovarian CCCs and is apromising therapeutic target in aggressive forms of ovariancancer.

Implications:AKT-targeted therapy has value in a first-line settingas well as a second-line treatment for recurrent disease developingafter platinum-based chemotherapy or bevacizumab treatment.Mol Cancer Res; 13(4); 795–806. �2014 AACR.

IntroductionOvarian carcinoma is the fourthmost common cause of cancer-

related deaths among women in the United States, with 22,240new cases diagnosed and approximately 14,030 deaths in 2013(1). Cytoreductive surgery followed by platinum-based chemo-therapy combined with paclitaxel has been the standard initialtreatment in patients with epithelial ovarian cancer (2). Recently,it has been reported that addition of bevacizumab during andafter carboplatin and paclitaxel chemotherapy prolongs progres-sion-free survival (PFS) by about 4 months in patients withadvanced epithelial ovarian cancer. (3, 4).However,many clinicalproblems still exist in the treatment of epithelial ovarian cancer.

One of the most important problems that needs to be resolvedis the management of clear cell carcinoma (CCC) of the ovary,which was first recognized by theWorld Health Organization as adistinct histologic subtype in 1973 (5). The major clinical pro-blems in the management of CCCs are its poor sensitivity to first-line platinum-based chemotherapy and the lack of effective

chemotherapy for recurrent CCCs (6). Therefore, to improvesurvival of patients with CCC, a better understanding of themechanism of platinum-resistance and the identification of effec-tive treatment strategies for both advanced and recurrent diseaseare needed.

The sensitivity of cancer cells to chemotherapeutic drug-induced apoptosis depends on the balance between proapoptoticand antiapoptotic signals (7, 8). Therefore, inhibition of anti-apoptotic signals, such as thosemediated by theAKTpathway, hasbeen proposed as a promising strategy to enhance the efficacy ofconventional chemotherapeutic agents (7, 8).

AKT is a serine–threonine protein kinase that has a crucial rolein cellular processes including glucose metabolism, apoptosis,and cell proliferation (7, 8). AKT is known to be activated by bothphosphoinositide-dependent kinase 1 (PDK1) and mTORC2,and activated AKT, in turn, phosphorylates multiple downstreamtargets via its kinase activity (7, 8). It has been reported that AKT isfrequently activated in epithelial ovarian cancer (9, 10). More-over, AKT inhibition therapy has been shown to inhibit theproliferation of ovarian cancer cells with hyperactivation of AKTboth in vitro and in vivo (10–12). However, because most tumorspecimens and tumor-derived cell lines used in these prior inves-tigations have been ovarian serous adenocarcinomas (SAC), therole of AKT in CCC remains largely unknown.

It has been reported that activating mutations of PIK3CA occurin about 40%ofovarianCCCs,which ismore frequent than in anyother histologic subtype of epithelial ovarian cancer (13). It hasalso been reported that loss of PTEN expression is common inCCC of the ovary (14). Moreover, it has also been recently

1Department of Obstetrics and Gynecology, Osaka University Gradu-ate School of Medicine, Osaka, Japan. 2Cancer Biology Program, FoxChase Cancer Center, Philadelphia, Pennsylvania.

Corresponding Author: Seiji Mabuchi, Department of Obstetrics and Gynecol-ogy, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita,Osaka 565-0871, Japan. Phone: 81-6-6879-3354; Fax: 81-6-6879-3359; E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-14-0314

�2014 American Association for Cancer Research.

MolecularCancerResearch

www.aacrjournals.org 795

on August 27, 2020. © 2015 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst December 17, 2014; DOI: 10.1158/1541-7786.MCR-14-0314

http://mcr.aacrjournals.org/

ACCC SAC

P-AKT (Ser 473) expression

Early stage

Advanced stage

(stage I–II)

(stage III–IV)

(i)

Exp

ress

ion

rate

of p

hosp

ho-A

KT

(%

) N.S.N.S.

CCC

SAC

Stage I–II Stage III–IV0

20

10

80

70

60

50

40

30

A (ii)

P-AKT(Ser473)

AKT

β-Actin

β-Actin

β-Actin

P-S6K1(Thr389)

S6K1

CRMG1 RMG2

0 3 10 30 500 3 10 30 50Perifosine (μmol/L)

E (ii)(i)

RM

G2

OV

ISE

SK

OV

3

A27

80

P-AKT(Ser473)

AKT

E

RM

G1

RM

G2

KO

C7C

HA

C2

P-AKT(Ser473)

AKT

B

D

Via

bilit

y ba

sed

on M

TS

ass

ay

Perifosine concentrations (μmol/L)

120

100

80

60

40

20

01 3010

RMG1

RMG2

KOC7C

HAC2

Via

bilit

y ba

sed

on M

TS

ass

ay


120

100

80

60

40

20

01 3010

RMG2

OVISE

SKOV3

A2780

Figure 1.AKT is frequently activated and can be a therapeutic target in ovarian CCCs. A, i, ovarian cancer tissue microarrays were stained with phospho-AKT (Ser473)antibody, and representative photographs of ovarian tissue microarray cores are shown. (Continued on the following page.)

Sasano et al.

Mol Cancer Res; 13(4) April 2014 Molecular Cancer Research796




reported that mTORC2 is activated in approximately 70% ofCCCs (15). Because these genetic and epigenetic changes resultsin the hyperactivation of AKT signaling, CCCs may be morestrongly dependent on AKT signaling for tumor progression thanare other histologic subtypes of epithelial ovarian cancer, and thusAKT may be a very promising therapeutic target in CCC. Giventhat patients with CCC have poor prognosis, hopes are high forthe development of AKT-targeting therapy in this patientpopulation.

Perifosine is a synthetic alkylphospholipid that inhibits theactivation of AKT through preventing cell membrane recruitmentof the N-terminal AKT pleckstrin homology (PH) domain (16).Previous studies with perifosine demonstrated antitumor activi-ties in multiple human trials (16). Perifosine has also shownsignificant antitumor activity either as a single agent (17) or incombination with paclitaxel (18) in preclinical studies ovariancancer. However, the activity of perifosine in CCC remainsunknown.

In this study, we examined the activation status of AKT both inearly-stage and advanced-stage CCC, and determined whetherperifosine has antineoplastic efficacy in both in vitro and in vivomodels of CCC. Moreover, we investigated the potential role ofAKT-inhibition therapy in CCCs that had acquired resistance aftertreatment with cisplatin or bevacizumab treatments.

Materials and MethodsReagents/antibodies

Perifoine was obtained from Aeterna Zentaris GmbH. Antibo-dies recognizing AKT, phospho-AKT (Ser473), S6K1, phospho-S6K1 (Thr389), poly(ADP ribose) polymerase (PARP) andb-actin, were obtained from Cell Signaling Technology. Anti-rabbit secondary antibodies were purchased from Santa CruzBiotechnology. The CellTiter 96-well proliferation assay kitswere obtained from Promega. Cisplatin was purchased fromSigma. Bevacizumab was kindly provided by Chugai Pharmaceu-tical Co., Ltd.

Cell lines and culturesHuman ovarian CCC cell lines, RMG1, RMG2, KOC7C, and

HAC2, were kindly provided by Dr. H. Itamochi (Tottori Univer-sity, Tottori, Japan).HumanovarianCCC cell line,OVISE, humanSAC cell lines, SKOV-3 and A2780, were purchased from theAmerican Type Culture Collection. Human ovarian adenocarci-noma cell lines, OVCAR4 and OVCAR5, were kindly provided byCell Culture Facility at Fox Chase Cancer Center (Philadelphia,PA). We tested these cells lines in our laboratory for their authen-tication by morphologic observation. No further cell line authen-tication was conducted by the authors. Each cell line was never

continuously passaged in culture for more than 3 months, andafter that, a new vial of frozen cells was thawed. Cells werecultured in DMEM/Ham's F-12 (Gibco) with 10% fetal bovineserum, as reported previously (15).

Establishment of cisplatin-resistant cell linesCisplatin-resistant sublines fromRMG1 and RMG2were devel-

oped by continuous exposure to cisplatin as described previously(19). Briefly, cells of both lines were exposed to stepwise increasesin cisplatin concentrations. Initial cisplatin exposure was at aconcentration of 1 mmol/L. After the cells had regained theirexponential growth rate, the cisplatin concentration was doubledand then the procedure was repeated until selection at 10 mmol/Lwas attained. The resulting cisplatin-resistant sublines, designatedas RMG1-CR and RMG2-CR, were cultured inDMEM/Ham's F-12containing 10 mmol/L cisplatin to maintain a high level ofcisplatin resistance.

Cell proliferation assayTheMTS assay was used to analyze the effect of perifosine and/

or cisplatin on cell viability as described previously (20). Cellswere cultured overnight in 96-well plates (1� 104 cells/well). Cellviability was assessed 48 hours after addition of perifosine or/andcisplatin at the indicated concentrations. The number of survivingcells was assessed by determination of theA490 nm of the dissolvedformazan product after addition ofMTS for 1 hour as described bythemanufacturer (Promega). Cell viability is expressed as follows:Aexp group/Acontrol � 100.

Clonogenic survival assayThe cells were plated into 6-cm dishes and then perifosine was

added for 24 hours. All of the cells were trypsinized and counted.The cells (5� 102/well) were plated in 6-well tissue culture platesin medium containing 10% fetal bovine serum. The plates wereincubated for 10 to 14 days. Cells were fixed and stained usingDiff-Quik (DadeBehring), and the number of colonies, consistingof�50 cells, in triplicate wells was counted as surviving colonies.

Cell-cycle analysisCells were incubated with perifosine at the indicated concen-

trations for 24 hours. Cells were then fixed with 75% ethanolovernight at 4�C, and stained with propidium iodide (PI; 50 mg/mL) in the presence of RNase A (100 mg/mL; Roth) for 20minutesat 4�C. Cell-cycle distribution was determined by analyzing10,000 cells using a FACScan flow cytometer and Cell Questsoftware (Becton Dickinson) as reported previously (15).

(Continued.) Magnifications: �100 and �400 (insets). ii, Histogram indicating the frequency of phospho-AKT staining by clinical stage and histologic type. N.S.,no statistically significant difference. B, AKT activation status in four ovarian CCC cell lines. CCC cells were incubated in the presence of 10% FBS, after whichthe AKT activity was determined by Western blotting. Actin expression was used as a loading control. C, the effect of perifosine on AKT-mTOR activation in vitro.RMG1 and RMG2 cells were treated with 0, 3, 10, 30, or 50 mmol/L perifosine for 6 hours in the presence of 10% FBS. Cells were harvested, and equivalent amounts(30 mg) of protein were subjected to SDS-PAGE and blotted with anti-phospho-AKT (Ser473), anti-AKT, anti-phospho-S6K1 (Thr389), anti-S6K1, or anti-b-actinantibodies. D, sensitivity of CCC cells to perifosine. SAC cells were treated with the indicated concentrations of perifosine in the presence of 10% FBS for 72 hours.Cell viability was assessed by the MTS assay. The experiment was performed six times. Data are shown as the mean of six experiments. E, i, sensitivity of SAC cellsand CCC cells to perifosine. CCC cells (OVISE and RMG2) and SAC cells (SKOV-3 and A2780) were treated with the indicated concentrations of perifosinein thepresenceof 10%FBS for 48 hours. Cell viabilitywas assessedby theMTS assay. The experimentwasperformed three times. Data are shownas themeanof threeexperiments. ii, AKT activation status in four cell lines. CCC cells were incubated in the presence of 10% FBS, after which the AKT activity was determined byWestern blotting. Actin expression was used as a loading control.

AKT-Targeted Therapy for Ovarian CCC

www.aacrjournals.org Mol Cancer Res; 13(4) April 2014 797




Western blottingCells treated as indicated were lysed for 10 minutes at 4�C.

Equal amounts of proteins were separated by SDS-PAGE andtransferred to nitrocellulose membranes. Blocking was done in5% nonfat milk in 1� Tris-buffered saline. Western blot analyseswere performed using specific primary antibodies. Immunoblotswere visualized with horseradish peroxidase–coupled immuno-globulin by using an enhanced chemiluminescence Westernblotting system (PerkinElmer).

In vitro VEGF protein quantitationRMG2 cells (2 � 106) were incubated in DMEM Ham's F-12

medium containing 10% FBS for 24 hours under normoxia orhypoxia (1% O2). Then, the culture supernatants were collectedand levels of VEGFwere determined using theQuantikineHumanVascular Endothelial Growth Factor Immunoassay (R&D Sys-tems) according to the manufacturer's protocol. The remainingmonolayers were trypsinized, and the cells were counted tonormalize VEGF protein values. VEGF values were derived from

A

RMG1

RMG2

0.2%Sub-G1

G1

S

G2–M

13.3%

27.2%

59.3%

9.7%Sub-G1

G1

S

G2–M

9.4%

11.2%

69.7%

0.2%Sub-G1

G1

S

G2–M

9.3%

20.7%

69.8%

10.8%Sub-G1

G1

S

G2–M

7.3%

9.6%

72.3%

Control Perifosine 30 μmol/L

B

β-Actin

PARP

RMG1

Perifosine (μmol/L) 0 3 10 30 50

(i)

B (ii)


Apo

ptos

is c

ell r

atio

(%

)

Early apoptosis

Total apoptosis

Late apoptosis

0

10

40

20

30


RMG1 RMG2

**

*

*

*

*

*

RMG2

0 3 10 30 50

Figure 2.Perifosine induces both cell-cycle arrestand apoptosis. A, perifosine induces bothcell-cycle arrest in G1-phase. RMG1 andRMG2 cells were treated with or without30 mmol/L perifosine for 24 hours. Cell-cycle analyses were performed with afluorescence-activated cell sorter (FACS)can as described in the Materials andMethods. B, perifosine induces apoptosis.RMG1 and RMG2 cells were treated withor without 30 mmol/L perifosine for 24hours. Cells were harvested, and theeffect of perifosine was evaluated byWestern blotting (i) or flow cytometry (ii)as described in Materials and Methods.

Sasano et al.





a standard curve of known concentrations of recombinant humanVEGF. Each sample was analyzed in duplicate and averaged.

Tube formation assayTube formation assay was done as described previously (21).

The surfaces of 96-well plates were coated with 30 mL of growthfactor–reduced Matrigel matrix (BD Biosciences). Then, 1� 104

serum-starved human umbilical vein endothelial cells(HUVEC) in 100 mL of M199 medium containing 0.5% bovineserum albumin were plated. Perifosine was added at the time ofplating. After 8-hour incubation, tube formation was visualizedunder an inverted microscope (�40) and the images wereanalyzed.

HUVEC proliferation assaySerum-starved HUVECs were plated at 1� 104 cells per 96 well

with perifosine at the indicated concentrations. The cell wereincubated at 37�C and 5% CO2 for 24 hours in HuMedia-EG2supplemented with 2% fetal bovine serum (Kurabo Industries)and proliferation was assessed by the MTS assay (Promega).

Detection of apoptosisCCC cells were treated with 30 mmol/L of perifosine for 24

hours. Then, cells were harvested, and stained with PI and annexinV using the annexin V–FITC apoptosis detection kit (BioVision),according to the manufacturer's instructions. Fluorescence datawere collected using flow cytometry. The sum total of early apop-topic cells, annexinV(þ) PI(�), and late apoptopic cells, annexinV(þ) PI (þ), was defined as the total number of apoptotic cells.

Clone selectionThe plasmid encoding constitutively active AKT2 (HA-Myr-

AKT2) or the control vector (pcDNA3) used in this study havebeen described previously (22). The plasmids encoding thekinase-dead AKT (HA-AktK179M) and control vector (pCMV6)have been described previously (11, 23). RMG1 and RMG2cells were transfected in 6-well tissue culture plates with 1 mg ofthe pcDNA3, HA-Myr-AKT2, pCMV6, or HA-AktK179M usingLipofectamine 2000 according to the manufacturer's instruc-tions (Invitrogen). Clonal selection was performed by addingG418 (Famingdale, Enzo Life Sciences). The resulting stabletransfectants expressing control plasmid or HA-Myr-AKT2 weredesignated as RMG1-control, RMG1-HA-Myr-AKT2, RMG1-HA-AKT-KM, RMG2-control, RMG2-HA-Myr-AKT2, and RMG2-HA-AKT-KM.

Clinical samplesAll surgical specimens were collected and archived according to

protocols approved by the Institutional Review Boards of theparent institutions. Appropriate informed consent was obtainedfromeachpatient. The tumors included52CCCs and46SACs.On

the basis of criteria of the international Federation of Gynecologyand Obstetrics (FIGO), 27 CCCs were stages I–II tumors and 25were stages III–IV tumors. Among SACs, 22 were stages I–IItumors and 24 were stages III–IV tumors. Tumor samples werefixed in 10% neutral buffered formalin overnight and thenembedded in paraffin. Ovarian cancer tissue microarrays consist-ing of two cores from each tumor sample were prepared, asdescribed previously (15, 19).

ImmunohistochemistryTissue sections were cut at 4 mm, mounted on slides, and

processed for immunohistochemical staining. Sectionswere incu-bated with the primary antibody, followed by the appropriateperoxidase-conjugated secondary antibody. The slides werescored semiquantitatively by two pathologists who were blindedto the clinical outcome. Surrounding non-neoplastic stromaserved as an internal negative control for each slide. A score of0 indicated no staining, þ0.5 was weak focal staining (<10% ofthe cells stained), þ1 was indicative of focal positive staining(10%–50% of the cells stained), þ2 indicated clearly positivestaining (>50% of the cells stained), and a score of þ3 wasintensely positive. Tumors with staining of þ1, þ2, or þ3 com-prised the positive-staining groups. When the two cores from thesame tumor sample showed different positively results, the lowerscore was considered valid.

In vivo tumor studiesAll procedures involving animals and their care were approved

by the Institutional Animal Care and Usage Committee of OsakaUniversity (Osaka, Japan), in accordance with institutional andNIH guidelines.

Initial experiments were conducted to examine the antitumoractivity of perifosinemonotherapy inCCC. Then, 5- to 7-week-oldnude mice (n¼ 10) were inoculated s.c. into the right flank eitherwith 5� 106 RMG2 cells in 200 mL of PBS. When tumors reachedabout 50mm3, themicewere assigned into two treatment groups:placebo (n ¼ 5) or perifosine (n ¼ 5).

The second set of experiments was conducted to examine theantitumor effect of perifosine in CCC that had acquired resistanceto bevacizumab. To confirm the antitumor effect of bevacizumabon the growth of CCCs, 5- to 7-week-old nude mice (n¼ 10) wereinoculated s.c. into the rightflank either with 5� 106 RMG2 cells in200 mL of PBS. When tumors reached about 50 mm3, mice wereassigned into two treatment groups: bevacizumab (n ¼ 5) orplacebo (n ¼ 5). To confirm the antitumor effect of perifosine onCCCs that had acquired resistance to bevacizumab treatment, 5- to7-week-old nude mice (n ¼ 10) were inoculated s.c. into the rightflank either with 5 � 106 RMG2 cells in 200 mL of PBS. Whentumors reached about 50 mm3, all mice were treated with bev-acizumab. After 3 weeks of treatment, the mice were assigned intotwo treatment groups: bevacizumab continuation (n ¼ 5) or

Table 1. Effects of perifosine on cell-cycle progression

Sub-G1 G1 S G2–MCell lines Treatments % (SD) % (SD) % (SD) % (SD)

RMG1 Control 0.2 (�0.06) 59.0 (�0.30) 13.8 (�0.46) 27.0 (�0.32)Perifosine, 30 mmol/L 9.2b (�0.76) 71.3b (�2.18) 9.3 (�1.00) 10.2 (�1.50)

RMG2 Control 0.2 (�0.06) 70.6 (�0.66) 9.1 (�0.18) 20.1 (�0.46)Perifosine, 30 mmol/L 10.6b (�1.56) 74.1a (�1.70) 6.8 (�0.50) 8.5 (�2.18)

aP < 0.05, as compared with control.bP < 0.01, as compared with control.






RM

G1

OV

CA

R4

OV

CA

R5

P-AKT(Ser473)

AKT

β-Actin Via

bilit

y ba

sed

on M

TS

ass

ay


120

100

80

60

40

20

1 3010

OVCAR4

OVCAR5

RMG1

A (i) A (ii)

0

***

**

RM

G1-

cont

rol

RM

G1-

HA

-Myr

- AK

T2

P-AKT(Ser473)

AKT

β-Actin

B

C

Via

bilit

y ba

sed

on M

TS

ass

ay

**

Control

Cisplatin 30 μmol/L

120

100

80

60

40

20

0

(i) C (ii)

Via

bilit

y ba

sed

on M

TS

ass

ay


RMG1

***

120

100

80

60

40

20

0

RMG1-HA-Myr-AKT2

1 10 30

RMG1-control

RM

G2-

cont

rol

RM

G2-

HA

-Myr

- AK

T2

**

D

RM

G1-

HA

-AK

T-K

M

P-AKT(Ser473)

AKT

β-Actin

(i)

RM

G1-

cont

rol

RM

G2-

HA

- AK

T-K

M

RM

G2-

cont

rol

Via

bilit

y ba

sed

on M

TS

ass

ay

120

100

80

60

40

20

0

Control

Perifosine 30 μmol/L

**

***

*

** *

D (ii)V

iabi

lity

base

d on

MT

S a

ssay


RMG2120

100

80

60

40

20

0

RMG2-HA-Myr-AKT2

1 10 30

RMG2-control

****

Figure 3.AKT activation as a predictor of perifosine sensitivity. A, sensitivity to perifosine according to AKT activation status. Ovarian cancer cells with differential AKT activationwere treated with the indicated concentrations of perifosine and incubated in medium containing 10% FBS for 48 hours. Cell viability was assessed by the MTSassay as described in theMaterials andMethods. i, AKT activation status. ii, Cell viabilitywas assessed by theMTS assay. The experimentwas performed six times. Data areshown as the mean of six experiments. B, activation status in CCC cells after transfection with control vector (pcDNA3) or constitutively active AKT2 (HA-Myr-AKT2).Note that in B, the cells were harvested after serum-starvation overnight. Western blotting was carried out with anti-phospho-AKT (Ser473), anti-AKT, oranti-b-actin antibody. C, sensitivity to cisplatin (i) or perifosine (ii) according to AKT activation status. CCC cells stably transfected with control vector (pcDNA3) orconstitutively active AKT2 (HA-Myr-AKT2) were treated with the indicated concentrations of perifosine and incubated in medium containing 10% FBS for 48 hours. Cellviability was assessed by the MTS assay. D, i, activation status in CCC cells after transfection with control vector (pCMV6) or kinase-dead AKT (HA-AktK179M). ii,Sensitivity toperifosineaccording toAKTactivationstatus.CCCcells stably transfectedwith controlvector (pCMV6)orkinase-deadAKT(HA-AktK179M)were treatedwiththe indicated concentrations of perifosine and incubated inmedium containing 10% FBS for 48 hours. Cell viability was assessed by theMTS assay. � , P < 0.05, �� , P < 0.01.

Sasano et al.





perifosine treatment (n¼ 5). Perifosine was administered intragas-trically using an animal-feeding needle. Perifosine was given in aloading dose of 75mg/kg (2� 37.5 mg/kg separated by 12 hours)followed by daily maintenance dose of 25 mg/kg for 14 days.Bevacizumab was administered intraperitoneally twice-weekly at adose of 5 mg/kg. Caliper measurements of the longest peripendi-cular tumor diameters were done every week to estimate tumorvolumeusing the following formula:V¼ L�W�D�p/6,whereVis the volume, L is the length, W is the width, and D is the depth.

Statistical analysisThe effect of AKT inhibition on cell proliferation and apoptosis

was analyzed by the Student t test. Tumor volumewas analyzed bythe Student t test and the Wilcoxon exact test. Immunoreactivitywas analyzed using the Fisher exact test.

ResultsAKT is frequently activated in CCCs

To determine the activation of AKT, tissue microarrays consist-ing of 52 ovarian CCCs and 46 ovarian SACs were examinedimmunohistochemically for phospho-AKT (Ser473). As shownin Fig. 1A, phospho-AKT (Ser473) expression was observed in68% of advanced-stage CCCs and in 70% of early-stage CCCs,compared with 66% and 49% in advanced-stage and early-stageSACs, respectively. In earlier stage (stage I–II), although thefrequency of phospho-AKT expression is slightly higher in CCCsand in SACs, the difference was not statistically significant.

In vitro growth-inhibitory effect of perifosine on CCC cell linesGiven the frequent AKT activation found in humanCCC tumor

specimens (Fig. 1A), we evaluated the activation of AKT in four

A

Tum

or g

row

th (m

m3 )

Days after treatment

RMG2

***

Control

300

250

200

150

100

0

50

0 4 7 11 14

BControl Perifosine

P-AKT

C

VE

GF

conc

entra

tion

(pg/

mL)

Perifosine (μmol/L) 0 300

0

1,400

1,200

1,000

800

600

400

200

* **

*

Hypoxia – ++

30

–

(i)

Via

bilit

y ba

sed

on M

TS a

ssay


0

20

40

60

80

100

120**

Control

(ii)


Per

cent

age

of c

ontro

l

0

20

40

60

80

100

120*

D

Tube

leng

th

Control

D

E (i) Control Perifosine

CD31

E (ii)

Mic

rove

ssel

are

a (M

VA

) (%

) 8

4

0Control Perifosine

**

Perifosine

Figure 4.Effect of perifosine on the growth ofCCC-derived tumor cells in vivo. A,graphdepictingweekly tumorvolumes.Athymic nudemicewere inoculated s.c.with RMG2 cells. When the tumorsreached an average size of about 50mm3, mice were treated with placeboor perifosine for 2 weeks. Points, mean;bars, SD. � , P < 0.05; �� , P < 0.01significantly different from placebo-treated mice. B, effect of perifosine onthe expression of phospho-AKT in vivo.Representative subcutaneous tumorsfrom placebo- or perifosine-treatedmice were excised, and stained withanti-phospho-AKT antibody. C, RMG2cells were treated with indicatedconcentrations of perifosine andincubated for 24 hours undernonhypoxic (20%O2) or hypoxic (1%O2

) conditions. The conditioned mediumwas collected, and the VEGFconcentrations were determined byELISA. D, i, antiproliferative activity ofperifosine on HUVEC in vitro. HUVECwere cultured with indicatedconcentration of perifosine in thepresence of 2% FBS for 24 hours. Cellviability was assessed by the MTSassay. ii, antiangiogenic activity ofperifosine determined using the tubeformation assay. Three random fieldsper samplewererecorded, and the tubelength of every field was measured.Each experiment was performed atleast three times, and data from onerepresentative experiment are shown.E, effect of perifosine on tumorangiogenesis. i, serial sections ofsubcutaneous tumors were stainedwith anti-CD31 antibody, andrepresentativephotographs are shown.Note the decreased microvesselstaining (dark brown) in therepresentative perifosine-treatedtumor compared with representativeplacebo-treated tumor. ii, A box blotindicating CD-31–positive microvesselarea (MVA) of placebo- or perifosine-treated tumors. Columns, mean; bars,SD. �� , P < 0.01.






Via

bilit

y ba

sed

on M

TS

ass

ay

Cisplatin concentrations (μmol/L)

RMG1

**

**

**

**

RMG1

RMG1-CR

120

100

80

60

40

20

01 3010

A

RMG2

RMG2-CR

P-AKT(Ser473)

AKT

β-Actin

CR

Par

enta

l

RMG1 RMG2BRMG1-CR

P-AKT(Ser473)

AKT

β-Actin

Perifosine (μmol/L)

RMG2-CR

C

0 3 10 30 50 0 3 10 30 50

Via

bilit

y ba

sed

on M

TS

ass

ay


RMG2120

100

80

60

40

20

01 3010

CR

Par

enta

l

Via

bilit

y ba

sed

on M

TS

ass

ay


**

**

***

RMG1120

100

80

60

40

20

01 3010

RMG1RMG1-CR

Via

bilit

y ba

sed

on M

TS

ass

ay


**

***

**

RMG2120

100

80

60

40

20

0

1 3010

RMG2RMG2-CR

D (i)

Via

bilit

y ba

sed

on M

TS

ass

ay


RMG1-CR

**

****

120

100

80

60

40

20

0



1 3010

E

Via

bilit

y ba

sed

on M

TS

ass

ay


RMG2-CR120

100

80

60

40

20

0



1 3010

**

D

Control


RMG1 RMG2RMG1-CR RMG2-CR

**

Rel

ativ

e co

lony

num

ber

1.2

1.0

0.8

0.6

0.4

0.2

0

**(ii)

Figure 5.Effect of perifosine on cisplatin-resistant CCC. A, establishment of cisplatin-resistant variant cell lines. Cisplatin-resistant sublines were established as describedin Materials and Methods. Cisplatin-sensitive parental (RMG1 and RMG2) and cisplatin-resistant variant (RMG1-CR and RMG2-CR) cells were treated with theindicated concentrations of cisplatin in the presence of 10% FBS for 48 hours. Cell viability was assessed by the MTS assay. Points, mean; bars, SD (�� , P < 0.01).B, activation of AKT in cisplatin-sensitive parental and cisplatin-resistant variant cells in vitro. RMG1, RMG1-CR, RMG2, and RMG2-CR cells were serum-starvedovernight. Cells were harvested, and equivalent amounts (30 mg) of proteinwere subjected to SDS-PAGE and blottedwith anti-phospho-AKT (Ser473), anti-AKT, oranti-b-actin antibodies. C, perifosine attenuates the phosphorylation of AKT in cisplatin-resistant CCC cells. (Continued on the following page.)


Sasano et al.




human CCC cell lines. As shown in Fig. 1B, AKT activation isobserved in all CCC cell lines tested, which is consistent withimmunohistochemical results observed with tumor samples.

Using these cell lines, we examined the antitumor effect of AKT-targeting therapy in vitro. For this purpose, we used the AKTinhibitor perifosine, which has been shown to inhibit the activityof AKT by preventing cell membrane recruitment of the AKT. Wefirst confirmed the effect perifosine on the activation of AKTsignaling in CCC cells. As shown, treatment of RMG1 and RMG2cells with perifosine significantly attenuated the phosphorylationof AKT and S6K1 in a dose-dependent manner, indicating thattreatment with perifosine effectively inhibited the AKT-mTORC1signaling pathway (Fig. 1C). Moreover, perifosine treatmentsignificantly inhibited the proliferation of all CCC cells tested tothe same extent (Fig. 1D). To investigate whether the growth-inhibitory effect of perifosine is specific for CCC, we next exam-ined the antiproliferative effect of perifosine using two CCC celllines and two SAC cell lines that express similar levels of phospho-AKT (Fig. 1E). As shown, these four cell lines showed similarsensitivity to perifosine, indicating that the perifosine has anti-tumor effect for ovarian cancers exhibiting activation of AKTirrespective of tumor histology.

To investigate the mechanism by which perifosine inhibits theproliferation of CCC cells, we first examined the effect of perifo-sine on cell-cycle progression by flow cytometry. As shown,treatment of both RMG1 and RMG2 cells with 30 mmol/L peri-fosine resulted in an increase in the percentage of cells inG1-phase(Fig. 2A and Table 1). Moreover, the percentage of apoptotic cellsin sub-G1 was also significantly increased after treatment withperifosine in both cell lines, which is consistent with Westernblotting (Fig. 2B, i) or a flow cytometry (Fig. 2B, ii and Table 1).Collectively, these results suggest that perifosine inhibits prolif-eration of CCC cells by inducing both cell-cycle arrest andapoptosis.

AKT activation as a predictor of perifosine sensitivityIt has been previously reported that the basal levels of phos-

phorylation of AKT correlate with sensitivity of cancer cells toPI3K–AKT inhibitors (7–12). Thus, we investigated whether AKTactivity is associatedwith the sensitivity of CCC cells to perifosine.As shown in Fig. 3A, a clear differential sensitivity was demon-strated depending on the phospho-AKT status. OVCAR4 andOVCAR5 that have lowAKT activity were insensitive to perifosine,versus the RMG1 cells with AKT activity showed high sensitivity toperifosine. To further investigate the potential of phospho-AKTexpression as a biomarker to predict the sensitivity to perifosine,wenext establishedCCC cell lines stably transfected constitutivelyactive AKT2, which resulted in constitutive phosphorylation ofAKT under serum-starvation conditions (Fig. 3B). As predicted,CCC cell lines stably transfected with HA-Myr-AKT2 showedrelatively lower sensitivity to cisplatin compared with thosetransfected with empty vector (pcDNA3; Fig. 3C, i). However, as

shown, overexpression of HA-Myr-AKT2 resulted in higher sen-sitivity to perifosine (Fig. 3C, ii). Moreover, inhibition of AKT bythe overexpression of kinase-dead AKT decreased the AKT phos-phorylation levels, and resulted in decreased sensitivity to peri-fosine (Fig. 3D). These results suggest that AKT activity may be abiomarker to predict sensitivity of CCC cells to perifosine. Theseresults suggest that AKT activity may be a biomarker to predictsensitivity of CCC cells to perifosine.

Perifosine inhibits tumor growth in a subcutaneous xenograftmodel

Wenext examined the growth-inhibitory effect of perifosine onovarian CCC cells in vivo. As shown in Fig. 4A, treatment withperifosine decreased tumor burden compared with placebo, indi-cating that perifosine has a marked antitumor activity, even whenused as a monotherapy. Importantly, drug treatment was welltolerated, with no apparent toxicity throughout the study. Toinvestigate the mechanism by which perifosine inhibits thegrowth of tumors derived from subcutaneously inoculatedhuman ovarian CCC cell, tumors harvested from the placebo-or perifosine-treated mice were evaluated immunohistochemi-cally. As shown in Fig. 4B, strong immunoreactivity for phos-phorylated AKT was detected in tumors from mice treatedwith placebo but was reduced in tumors from mice treated withperifosine. These data suggest that the growth-inhibitory effectof perifosine in vivo is associatedwith inhibition of AKT-signaling,confirming appropriate drug targeting in vivo. Given that the AKT–mTOR pathway is known to stimulate tumor-related angiogen-esis, we next examined the antiangiogenic activity of perifosineboth in vitro and in vivo. As shown, treatment of hypoxic RMG2cells with perifosine attenuated the expression of VEGF (Fig. 4C).Moreover, treatment with perifosine significantly inhibited theproliferation and the tube formation activity of theHUVEC in vitro(Fig. 4D).Consistentwith thesefindings, as shown inFig. 4E, largeCD31-immunopositive vessels were observed in tumors fromplacebo-treatedmice, whereas the CD31-immunopositive vesselswere smaller and fewer in tumors from perifosine-treated mice.Collectively, these results indicate that the antitumor effect ofperifosine is associated, at least in part, with inhibition of tumorangiogenesis.

Antitumor effect of perifosine on cisplatin-resistant celllines

Cisplatin resistance is regarded as a major clinical problem inthe management of CCC of the ovary, both in first-line andrecurrent settings (6). Relative platinum resistance of CCC cellscompared with SAC cells has been demonstrated in preclinicalstudies (24, 25). It has been previously reported that hyperactiva-tion of AKT is involved in the resistance of ovarian SAC cells tocisplatin (26); however, whether AKT is involved in resistance tocisplatin in CCC remains unknown. We then established cisplat-in-resistant sublines from RMG1 and RMG2 cells (Fig. 5A), and

(Continued.) RMG1-CR and RMG2-CR cells were treated with the indicated concentrations of perifosine for 6 hours in the presence of 10% FBS. Cells wereharvested, and equivalent amounts (30 mg) of protein were subjected to SDS-PAGE and blotted with anti-phospho-AKT (Ser473), anti-AKT, or anti-b-actinantibodies. D, enhanced sensitivity to perifosine in cisplatin-resistant CCC cells in vitro. Cisplatin-sensitive parental (RMG1 and RMG2) and cisplatin-resistantvariant (RMG1-CR and RMG2-CR) cells were treated with the indicated concentrations of perifosine in the presence of 10% FBS for 48 hours. Cell viability wasassessed by MTS (i) or by clonogenic survival assays (ii). Points, mean; bars, SD (� , P < 0.05; �� , P < 0.01). E, perifosine enhances the therapeutic efficacy of cisplatinin CCC cells. RMG1-CR and RMG2-CR cells were treated with various concentrations of cisplatin with or without 30 mmol/L perifosine in the presence of 10% FBS.Cell viability was assessed by the MTS assay. Points, mean; bars, SD (�, P < 0.05; �� , P < 0.01).






investigated the activity of AKT in both cisplatin-resistant sublinesand parental cisplatin-sensitive cells by Western blotting. Asshown in Fig. 5B, higher levels of phospho-AKT were observedinboth cisplatin-resistant cell lines comparedwith their respectiveparental cell lines.

Because increased AKT activity resulted in enhanced sensitivityof CCC cells to perifosine (Fig. 3C), we considered cisplatin-resistant sublines to be good candidates for treatment withperifosine. Thus, we next examined the inhibitory effect of peri-fosine on cisplatin-resistant and parental cisplatin-sensitive CCCcell lines. We first confirmed that treatment with perifosineeffectively inhibited phosphorylation of AKT in vitro (Fig. 5C).We next examined the inhibitory effect of perifosine on CCC cellviability by theMTS assay, which revealed a clear differential effectdependent on sensitivity to cisplatin (Fig. 5D). Cisplatin-resistantRMG1-CR and RMG2-CR cells were found to be significantlymore sensitive to perifosine than their respective parental celllines RMG1 and RMG2. Because RMG1-CR and RMG2-CR areinsensitive to cisplatin treatment (Fig. 5A), perifosine may holdpromise for the treatment for recurrent CCCs developing aftercisplatin treatment.

We also used RMG1-CR and RMG2-CR cells to determinewhether treatment with perifosine can sensitize RMG1-CR andRMG2-CR cells to cisplatin. As shown in Fig. 5E, in the presence of30 mmol/L of perifosine, the ability of cisplatin to inhibit cellproliferation and survival was significantly enhanced in both celllines.

Activity of perifosine on the growth of bevacizumab-resistantCCCs

On the basis of recent phase III clinical trials (3, 4), bevacizu-mab combined with carboplatin plus paclitaxel is becoming oneof the standard first-line treatments for patients with advancedepithelial ovarian cancer.However, in these studies, improvementin PFS by the addition of bevacizumab was only about 4 months(3, 4). Thus, development of improved treatments for recurrentcancers following bevacizumab treatment are urgently needed.Thus, we next decided to evaluate the antitumor efficacy ofperifosine on CCC xenografts following bevacizumab treatment.

As shown in Fig. 6A, treatment with bevacizumab significantlydecreased CCC tumor burden. However, when we continued thebevacizumab treatment, after roughly 3 weeks of exposure, thesubcutaneous tumors started to grow rapidly, presumably as aresult of acquired resistance to bevacizumab (Fig. 6A). Thus, in asubsequent experiment, after 3 weeks of bevacizumab treatment,mice were assigned into two treatment groups receiving eitherbevacizumab or perifosine alone. We found that when treatmentwas switched from bevacizumab to perifosine after the initialtreatement with bevacizumab for 3 weeks, growth of the subcu-taneous tumors was significantly inhibited (Fig. 6B). These resultsindicate that perifosinemayhave therapeutic efficacy for recurrentCCCs after bevacizumab treatment.

DiscussionDespite advances in platinum-based combination chemother-

apy, patients with CCC of the ovary, especially in advanced stageor recurrent disease, have a worse PFS and overall survival whencompared with patients with a serous histology (6). Thus, toimprove survival of ovarian CCC patients, novel treatments needto be developed.

One possible strategy to inhibit tumor progression and toenhance the efficacy of platinum-based chemotherapy in CCC isto target antiapoptotic signals, such as thosemediated by the AKTpathway. AKT is known to regulate various cellular pathways thatpromote cell survival, cell proliferation, angiogenesis, and inva-sion (7, 8). We and others have previously reported that AKT isfrequently activated in epithelial ovarian cancer and that inhibi-tion of AKT activity by PI3K inhibitors significantly inhibits cellproliferation (12) and enhances the activity of conventionalanticancer agents, including cisplatin (12, 26) and paclitaxel(11) in preclinical models of ovarian cancer. A more recentinvestigation suggested that phospho-AKT expression correlatedwith the overall survival and the platinum-response in ovariancancer patients (17). However, because most tumor specimensand tumor-derived cell lines used in these investigations havebeen ovarian SACs, the role of AKT as a therapeutic target in CCChas remained unknown.

A BT

umor

gro

wth

(m

m3 )


RMG2

Tum

or g

row

th(m

m3 )


RMG2

*

900

0

300400500

800

600700

0

200100

7 2814 21 35

PBSBEV

0 7 2814 21 350

300

400

500

600

200

100

BEV-BEVBEV-Perifosine

* * *

*

*

Figure 6.In vivo growth-inhibitory effect of perifosine on bevacizumab-resistant CCC. A, the effect of bevacizumab (BEV) on CCC growth in vivo. Athymic nude miceinoculated s.c. with RMG2 cellswere assigned into two treatment groups; bevacizumab (n¼ 5) or placebo (n¼ 5). B, the effect of perifosine onCCC that had acquiredresistance to bevacizumab. Athymic nude mice that were inoculated s.c. with RMG2 cells and received bevacizumab treatment for 3 weeks. Then, the mice wereassigned into two treatment groups; bevaizumab (n¼ 5) or perifosine (n¼ 5). Tumor volumesweremonitoredweekly, and growth curves of each treatment groupsare shown. Points, mean; bars, SD (� , P < 0.05).

Sasano et al.





The Cancer Genome Atlas (TCGA) Research Network hasrecently reported activation of the PI3K–AKT pathway in high-grade serous ovarian cancer in approximately 40% of thesetumors, mostly through somatic copy number alterations andnot through point mutations (27). In contrast, in ovarian CCC, aprevious study revealed activating mutations of the PIK3CA genein approximately 40% of cases (13), which is among the mostfrequent in solid malignancies of various origins. Moreover,previous investigations indicated that mTORC1 (a downstreameffector of AKT) and mTORC2 (an upstream stimulator of AKT)are frequently activated inCCCof the ovary, and in vitro and in vivostudies have suggested that both mTORC1 and mTORC2 arepromising therapeutic targets in CCC (19, 21, 25, 28). Collec-tively, these results strongly indicate that inhibition of the AKTpathway is a promising strategy for the clinical management ofovarian carcinomas of the CCC type.

Perifosine is an orally bioavailable AKT inhibitor that iscurrently being evaluated in phase I/II clinical studies in avariety of solid tumors (16), and this drug has shown encour-aging clinical activity when used in combination with capeci-tabine in patients with metastatic colorectal cancer (29). Inovarian cancer, a phase II trial designed to evaluate the efficacyof perifosine in combination with docetaxel was conducted inthe recurrent setting (30). Although satisfactory efficacy ofperifosine could not be demonstrated in this study, treatmentwith perifosine plus docetaxel appeared to be more effective inthose cases in which the PI3K–AKT pathway was mutationallyactivated (30), suggesting that clinical activity of perifosineshould be tested in patients whose tumors display activatedAKT signaling in future clinical trials.

CCC patients have a high frequency of PIK3CAmutations (13),mTORC2 activation (15), and loss of PTEN expression (14).Consistent with these previous findings, the frequent AKT activa-tion was observed in CCCs in this study: the expression of phos-pho-AKT was observed in both stages III–IV CCCs (68%) andstages I–II CCCs (70%). Consistent with previous findings usingother PI3K–AKT inhibitors (Fig. 2), perifosine treatment hadmorerobust antitumor activity in CCC cells with high AKT activity thanin cells with low AKT activity. Moreover, treatment of CCC cellswith perifosine significantly enhanced the antitumor effect ofcisplatin (Fig. 4E). Taken together, these findings indicate that AKTis a rational target for the treatment of CCCof the ovary.Moreover,our results indicate that an AKT inhibitor, such as perifosine, couldhave significant antitumor effects as a single agent or in combi-nation with cisplatin for both previously untreated CCCs andrecurrent CCCs developing after cisplatin treatment.

Perifosine has shown significant antitumor activity in a genet-ically engineered mouse model of ovarian endometrioid adeno-carcinomadisplaying hyperactivation of AKT (31). Perifosine alsoshowed significant antiproliferative activity and enhanced theactivity of cisplatin in cisplatin-resistant SAC cells with elevatedAKT activity in vitro (32). Moreover, in this study, CCC cells andSAC cells exhibiting similar levels of AKT activation showedsimilar sensitivity to perifosine (Fig. 1E). These results suggestthat ovarian cancers exhibiting hyperactivation of AKT wouldrepresent good candidates for AKT inhibition therapy irrespectiveof histologic subtypes.

An additional important finding of our study is the potentantitumor activity of perifosine in recurrent CCC followingbevacizumab treatment (Fig. 5C). Given that ovarian cancerpatients treated with bevacizumab in combination with stan-

dard platinum-based chemotherapy experienced disease pro-gression with a median PFS of 4 months in a phase III study(3, 4), it would be very important to explore the salvage treat-ment for recurrent tumors that develop after bevacizumabtreatment. Because the experimental in vivo model used in thisstudy mimics resistance development in bevacizumab-treatedpatients, our results suggest that AKT inhibitors might be effi-cacious for the clinicalmanagement of recurrent CCCs followingtreatment with bevacizumab.

The limitations of our studyneed tobe addressed. Althoughourin vitro investigations suggested that AKT is involved in thecisplatin resistance in CCC of the ovary (Fig. 5), due to the limitedclinical data, we could not investigate the association betweenimmunoreactivity for phospho-AKT and the tumor response toplatinum-based chemotherapies. Another potential weakness isour experimental design, subcutaneous xenograft model. Perito-neal dissemination is the main process of progression in humanovarian cancer, thus an orthotopic xenograft model of clear cellovarian cancer may accurately model advanced disease. Althoughwe tried to examine the antitumor effect of perifoine in anorthotopic model, the CCC cells lines used in this study did notdevelop ovarian tumors in nude mice (data not shown). Tovalidate the results from this study and make definitive conclu-sions, further investigations may be needed.

In summary, the results presented here indicate that AKT isfrequently activated in ovarian CCC and is a promising therapeu-tic target for this disease both as afirst-line therapy and as a salvagetreatment for recurrence after cisplatin or bevacizumab treatment.This work provides scientific rationale for future clinical trials ofAKT-targeting therapies in patients with ovarian CCC, a chemore-sistant histologic subtype characterized by frequent activation ofthe AKT pathway.

Disclosure of Potential Conflicts of InterestT. Kimura has received speakers bureau honoraria from Bayer Pharm Co.

Ltd., Takeda Pharm. Co. Ltd., and Yakuruto Pharm. Co. Ltd. No potentialconflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: T. Sasano, S. MabuchiDevelopment of methodology: T. Sasano, S. Mabuchi, K. HashimotoAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): T. Sasano, Y. Matsumoto, T. HisamatsuAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): T. SasanoWriting, review, and/or revision of the manuscript: T. Sasano, S. Mabuchi,M. Kawano, J.R. Testa, H. KurodaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): T. Sasano, S. Mabuchi, R. Takahashi, A. Isobe,T. KimuraStudy supervision: T. Sasano, S. Mabuchi, K. Sawada

Grant SupportThis work was supported by grant-in-aid for General Scientific Research,

from the Ministry of Education, Culture, Sports, Science and Technology ofJapan (no. 26462523). J.R. Testa received support from NCI Grant R01CA77429 and P30CA006927.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received June 3, 2014; revised November 15, 2014; accepted December 7,2014; published OnlineFirst December 17, 2014.






References1. Siegel R, NaishadhamD, Jemal A. Cancer statistics, 2013. CA Cancer J Clin

2013;63:11–30.2. Ozols RF, Bundy BN, Greer BE, Fowler JM, Clarke-Pearson D, Burger RA,

et al.Gynecologic Oncology Group. Phase III trial of carboplatin andpaclitaxel comparedwith cisplatin andpaclitaxel in patientswithoptimallyresected stage III ovarian cancer: a Gynecologic Oncology Group study.J Clin Oncol 2003;21:3194–200.

3. Burger RA, Brady MF, Bookman MA, Fleming GF, Monk BJ, Huang H,et al.Gynecologic Oncology Group. Incorporation of bevacizumab inthe primary treatment of ovarian cancer. N Engl J Med 2011;365:2473–83.

4. Perren TJ, Swart AM, Pfisterer J, Ledermann JA, Pujade-Lauraine E, Kris-tensen G, et al. ICON7 Investigators. A phase 3 trial of bevacizumab inovarian cancer. N Engl J Med 2011;365:2484–96.

5. Serov SF, Scully RE, Sobin LH. International histologic classification oftumors. In: Histologic typing of ovarian tumors, vol. 9. Geneva, Switzer-land: World Health Organization; 1973.

6. Del Carmen MG, Birrer M, Schorge JO. Clear cell carcinoma of the ovary: areview of the literature. Gynecol Oncol 2012;126:481–90.

7. Vivanco I, Sawyers CL. The phosphatidylinositol 3-kinase AKT pathway inhuman cancer. Nat Rev Cancer 2002;2:489–501.

8. Engelman JA. Targeting PI3K signalling in cancer: opportunities, challengesand limitations. Nat Rev Cancer 2009;9:550–62.

9. Cheng JQ, Godwin AK, Bellacosa A, Taguchi T, Franke TF, Hamilton TC,et al. AKT2, a putative oncogene encoding a member of a subfamily ofprotein-serine/threonine kinases, is amplified in human ovarian carcino-mas. Proc Natl Acad Sci U S A 1992;89:9267–71.

10. Altomare DA, Wang HQ, Skele KL, De Rienzo A, Klein-Szanto AJ, GodwinAK, et al. AKT andmTORphosphorylation is frequently detected in ovariancancer and can be targeted to disrupt ovarian tumor cell growth. Oncogene2004;23:5853–7.

11. Mabuchi S, Ohmichi M, Kimura A, Hisamoto K, Hayakawa J, Nishio Y,et al. Inhibition of phosphorylation of BAD and Raf-1 by Akt sensitizeshuman ovarian cancer cells to paclitaxel. J Biol Chem 2002;277:33490–500.

12. Ohta T, Ohmichi M, Hayasaka T, Mabuchi S, Saitoh M, Kawagoe J, et al.Inhibition of phosphatidylinositol 3-kinase increases efficacy of cisplatinin in vivo ovarian cancer models. Endocrinology 2006;147:1761–9.

13. Kuo KT, Mao TL, Jones S, Veras E, Ayhan A, Wang TL, et al. Frequentactivating mutations of PIK3CA in ovarian clear cell carcinoma. Am JPathol 2009;174:1597–601.

14. Hashiguchi Y, TsudaH, Inoue T, Berkowitz RS,Mok SC. PTENexpression inclear cell adenocarcinoma of the ovary. Gynecol Oncol 2006;101:71–5.

15. Hisamatsu T,Mabuchi S,Matsumoto Y, KawanoM, Sasano T, Takahashi R,et al. Potential role of mTORC2 as a therapeutic target in clear cellcarcinoma of the ovary. Mol Cancer Ther 2013;12:1367–77.

16. Gills JJ, Dennis PA. Perifosine: update on a novel Akt inhibitor. Curr OncolRep 2009;11:102–10.

17. Al Sawah E, Chen X, Marchion DC, Xiong Y, Ramirez IJ, Abbasi F, et al.Perifosine, an AKT inhibitor, modulates ovarian cancer cell line sensitivityto cisplatin-induced growth arrest. Gynecol Oncol 2013;131:207–12.

18. Sun H, Yu T, Li J. Co-administration of perifosine with paclitaxel syner-gistically induces apoptosis in ovarian cancer cells: more than just AKTinhibition. Cancer Lett 2011;310:118–28.

19. Mabuchi S, Kawase C, Altomare DA, Morishige K, Sawada K, Hayashi M,et al.mTOR is a promising therapeutic target both in cisplatin-sensitive andcisplatin-resistant clear cell carcinoma of the ovary. Clin Cancer Res2009;15:5404–13.

20. Mabuchi S, Ohmichi M, Kimura A, Ikebuchi Y, Hisamoto K, Arimoto-Ishida E, et al. Tamoxifen inhibits cell proliferation via mitogen-activatedprotein kinase cascades in human ovarian cancer cell lines in amanner notdependent on the expression of estrogen receptor or the sensitivity tocisplatin. Endocrinology 2004;145:1302–13.

21. Mabuchi S, Kawase C, Altomare DA, Morishige K, Hayashi M, Sawada K,et al. Vascular endothelial growth factor is a promising therapeutic targetfor the treatment of clear cell carcinoma of the ovary. Mol Cancer Ther2010;9:2411–22.

22. Tanno S, Tanno S, Mitsuuchi Y, Altomare DA, Xiao GH, Testa JR. AKTactivation up-regulates insulin-like growth factor I receptor expression andpromotes invasiveness of human pancreatic cancer cells. Cancer Res2001;61:589–93.

23. Datta SR, Dudek H, Tao X, Masters S, Fu H, Gotoh Y, et al. Akt phosphor-ylation of BAD couples survival signals to the cell-intrinsic death machin-ery. Cell 1997;91:231–41.

24. Itamochi H, Kigawa J, Sultana H, Iba T, Akeshima R, Kamazawa S, et al.Sensitivity to anticancer agents and resistance mechanisms in clear cellcarcinoma of the ovary. Jpn J Cancer Res 2002;93:723–8.

25. Mabuchi S, Hisamatsu T, Kawase C, HayashiM, Sawada K,Mimura K, et al.The activity of trabectedin as a single agent or in combination witheverolimus for clear cell carcinoma of the ovary. Clin Cancer Res2011;17:4462–73.

26. Mabuchi S, Ohmichi M, Nishio Y, Hayasaka T, Kimura A, Ohta T, et al.Inhibition of NFkappaB increases the efficacy of cisplatin in in vitro and invivo ovarian cancer models. J Biol Chem 2004;279:23477–85.

27. Cancer Genome Atlas Research Network. Integrated genomic analyses ofovarian carcinoma. Nature 2011;474:609–15.

28. Mabuchi S, Hisamatsu T, Kimura T. Targeting mTOR signaling pathway inovarian cancer. Curr Med Chem 2011;18:2960–8.

29. Bendell JC, Nemunaitis J, Vukelja SJ, Hagenstad C, Campos LT, HermannRC, et al. Randomized placebo-controlled phase II trial of perifosine pluscapecitabine as second- or third-line therapy in patients with metastaticcolorectal cancer. J Clin Oncol 2011;29:4394–400.

30. Fu S, Hennessy BT, Ng CS, Ju Z, Coombes KR,Wolf JK, et al. Perifosine plusdocetaxel in patients with platinum and taxane resistant or refractory high-grade epithelial ovarian cancer. Gynecol Oncol 2012;126:47–53.

31. WuR,HuTC, Rehemtulla A, Fearon ER,ChoKR. Preclinical testing of PI3K/AKT/mTOR signaling inhibitors in a mouse model of ovarian endome-trioid adenocarcinoma. Clin Cancer Res 2011;17:7359–72.

32. Engel JB, Sch€onhals T, H€ausler S, Krockenberger M, Schmidt M, Horn E,et al. Induction of programmed cell death by inhibition of AKT with thealkylphosphocholine perifosine in in vitro models of platinum sensitiveand resistant ovarian cancers. Arch Gynecol Obstet 2011;283:603–10.


Sasano et al.




2015;13:795-806. Published OnlineFirst December 17, 2014.Mol Cancer Res Tomoyuki Sasano, Seiji Mabuchi, Hiromasa Kuroda, et al. the OvaryPreclinical Efficacy for AKT Targeting in Clear Cell Carcinoma of

Updated version

10.1158/1541-7786.MCR-14-0314doi:

Access the most recent version of this article at:

Cited articles

http://mcr.aacrjournals.org/content/13/4/795.full#ref-list-1

This article cites 31 articles, 11 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/13/4/795.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://mcr.aacrjournals.org/content/13/4/795To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-14-0314

http://mcr.aacrjournals.org/content/13/4/795.full#ref-list-1

http://mcr.aacrjournals.org/content/13/4/795.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/13/4/795


PreclinicalEfﬁcacyforAKTTargetinginClearCell Carcinoma of ... · Carcinoma of the Ovary Tomoyuki...

Documents

Transcript of PreclinicalEfﬁcacyforAKTTargetinginClearCell Carcinoma of ... · Carcinoma of the Ovary Tomoyuki...