SU6656

Tyrosine Kinase Inhibitors as Potential Therapeutic Agents in the Treatment of Granulosa Cell Tumors of the Ovary

Stacey Jamieson, PhD and Peter J. Fuller, BMedSci, MBBS, PhD, FRACP

Objective: Granulosa cell tumors of the ovary (GCTs) represent a specific subset of ma- lignant ovarian tumors, of which there are 2 distinct subtypes, the juvenile and the adult form. Aside from surgery, no reliable therapeutic options currently exist for patients with GCT. This study sought to investigate the potential role of small molecule tyrosine kinase inhibitors (TKIs) as novel therapeutics in the clinical management of GCT.
Materials and Methods: Using TKI with distinct but overlapping multitargeted specific- ities, cellular proliferation, viability, and apoptosis were evaluated in 2 human GCT-derived cell lines, COV434 and KGN.
Results: Sunitinib, which targets the imatinib-inhibited tyrosine kinases of VEGFR, KIT, PDGFR, and FLT-3, was without effect in COV434 and KGN cell lines. Sorafenib, which has a high affinity for RAF1 and BRAF, dose dependently inhibited cellular proliferation and viability in both cell lines at concentrations equivalent to that seen in other systems. A RAF1 kinase inhibitor was without effect, suggesting that sorafenib is acting via inhibition of BRAF, or that aberrant signaling originates upstream of BRAF in the MAPK pathway. In the presence of a selective Src family inhibitor (SU6656), cell proliferation and cell viability responses dissociated; that is, although SU6656 dose dependently inhibited cell viability, it had limited effect on proliferation and apoptosis.
Conclusions: These findings implicate BRAF in the activated signaling responsible for the growth and viability of GCT and suggest that TKI already in clinical use may be a therapeutic option in the treatment of GCT.
Key Words: Granulosa cell tumor, Ovary, Molecular pathogenesis, Tyrosine kinase inhibitor, KGN, COV434
Abbreviations: ABL – Abelson murine leukemia viral oncogene homolog, AP-1 – activator protein1,ATP-adenosinetriphosphate,CML-chronicmyelogenousleukemia,EGFR-epidermal growth factor receptor, GCT- granulosa cell tumor of the ovary, HUVEC – human umbilical vein endothelial cell line, NFJB – nuclear factor JB, PDGFR – platelet-derived growth factor receptor, RTK – receptor tyrosine kinase, SCFR – Mast/stem cell growth factor receptor, TK- tyrosine kinase, TKI – tyrosine kinase inhibitor, VEGF – vascular endothelial growth factor, VEGFR – vascular endothelial growth factor receptor

Prince Henry’s Institute of Medical Research, and the Department of Medicine, Monash University, Clayton, Victoria, Australia.
Address correspondence and reprint requests to Peter J. Fuller, BMedSci, MBBS, PhD, FRACP, PHI Institute of Medical Research, 27-31 Wright St, Clayton, Victoria 3168, Australia. E-mail: [email protected].
Supported by grants-in-aid from the Cancer Council Victoria, the National Australia Bank Ovarian Cancer Research Foundation, the Granulosa Cell Tumor of the Ovary Foundation, and the National Health and Medical Research Council of Australia through a Senior Principal Research Fellowship to P.J.F.
(#1002559) and a Dora Lush Biomedical Postgraduate Research Scholarship to S.J. (#441132). S.J. was also in receipt of a Faculty of Medicine Postgraduate Excellence Award from Monash University. MIMR-PHI Institute of Medical Research is supported by the Victorian Government’s Medical Research Operational Infrastructure Support program. PHI Internal Data Audit #13Y24.
The authors declare no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal’s Web site (www.ijgc.net).

Copyright * 2015 by IGCS and ESGO
ISSN: 1048-891X
DOI: 10.1097/IGC.0000000000000479
Present address: Cancer Research UK Cambridge Institute, University of Cambridge, Robinson Way, Cambridge CB2 0RE, United Kingdom (Stacey Jamieson, PhD).

1224 International Journal of Gynecological Cancer & Volume 25, Number 7, September 2015

Received January 20, 2015, and in revised form April 12, 2015. Accepted for publication April 16, 2015.
(Int J Gynecol Cancer 2015;25: 1224Y1231)

ranulosa cell tumors of the ovary (GCTs) are thought to arise from normal proliferating granulosa cells of the late
preovulatory follicle and display many morphological and biochemical features of these cells.1,2 The finding that a so- matic missense mutation in the FOXL2 gene (c.402CYG; p.C134W) is present almost universally in adult GCT, although absent from juvenile GCT and other human malignancies,3Y12 is suggestive of an oncogenic or gain-of-function mutation and that the mutation is pathognomonic for the adult subtype.6,13
Granulosa cell tumors of the ovary are characterized by their slow, indolent progression and high rate of recurrence. Conventional treatment is predominantly surgical resection followed by radiotherapy or chemotherapy for recurrent or advanced disease. However, current chemotherapy is based on protocols derived largely from experience with ovarian epi- thelial tumors, and the efficacy of these interventions remains limited.14 As a result, approximately 80% of patients with ad- vanced stage or recurrent tumors succumb to their disease.13
Two human GCT-derived cell lines, COV434 and KGN, have been used as in vitro models to investigate the molec- ular pathogenesis of GCT. In light of the identification of the FOXL2 C134W mutation in adult GCT, the presence of the mutation in the KGN cell line9,15 and absence from the COV434 cell line6 suggests that these cell lines are derived from adult and juvenile GCT, respectively. Moreover, as well as being FOXL2 mutation negative, the COV434 cell line does not express the FOXL2 gene, providing further evi- dence that it represents a juvenile GCT of an advanced tumor stage.6,13 COV434 and KGN cell lines display constitutive nuclear factor JB (NFJB) and activator protein 1 (AP-1) transcriptional activity.16 Inhibition of NFJB activity using 2 independent inhibitors of IJB> phosphorylation results in a dose-dependent decrease in cell proliferation and cell via- bility, and a dose-dependent increase in apoptosis.14
Two casereportsofextensively pretreated, recurrentGCT responding to the tyrosine kinase inhibitor (TKI), imatinib,17,18 suggests a potential role for these therapies in the management of GCT. On this basis, Chu et al19 characterized the expres- sion of the imatinib-sensitive TK genes [Mast/stem cell growth factor receptor (SCFR)/KIT, Abelson murine leukemia viral oncogene homolog (ABL), platelet-derived growth factor re- ceptor (PDGFR)-> and -A] in human GCT samples, and ex- amined the effect of imatinib, and subsequently nilotinib, a more potent second-generation TKI with a profile similar to that of imatinib, on COV434 and KGN cell lines. All 4 kinases were expressed but at levels lower than those observed in pre- menopausal ovarian samples.19 Known activating mutations in KIT and PDGFRA were not found. Both cell lines responded to imatinib and nilotinib, showing dose-dependent decreases in cell proliferation and viability. These responses paralleled those observed in an imatinib-sensitive CML cell line (K562) but at
approximately 240- and 1000-fold higher concentrations of imatinib and nilotinib, respectively, suggesting that GCT are, in general, unlikely to respond to imatinib or nilotinib therapy.19 Theresponseimpliesan off-target effect, withimatinibinhibiting aTKother thanoneofthe4characterized.19 Astheconcentration of imatinib is increased, a range of other TKs is inhibited, in- cluding FMS-related tyrosine kinase 3 (FLT3; STK1), colony stimulating factor 1 receptor (CSF1R), SRC, and the EGFR.20 Given that overexpression of colony-stimulating factor (CSF) with its receptor (CSF1R) in normal granulosa cells resulted in proliferation and tumorigenesis,21 Chu et al19 also examined CSF1R and FLT3; their expression was either low or absent, makingitunlikelythattheyaremediatingtheresponse.Although these findings suggest that the TKs specifically targeted by therapeutic concentrations of imatinib and nilotinib do not play a pathogenic role in GCT, they argue that a TKI of appropriate specificity may represent a therapeutic option.19
As TKI are rationally designed to block the adenosine triphosphate (ATP)Ybinding pocket, at therapeutic doses, they exhibit a high affinity for their target kinase. However, be- cause TKs have a well-conserved ATP-binding site, there is a great potential for cross-reactivity, thereby eliciting a multitargeted effect.22 By using TKI with distinct but overlapping multitargeted specificities, and evaluating the end points of cell proliferation, cell viability, and apoptosis, the aim of this study was to identify a potential targeted therapeutic for the clinical treatment of GCT.

MATERIALS AND METHODS Cell Lines
The COV434 cell line, established in 1984 from a met- astatic GCT obtained from a 27-year-old patient,23 and the KGN cell line, established in 1994 from a recurrent, metastatic GCTremoved from a 73-year-old patient24 were used as invitro model systems of granulosa cell tumorigenesis. COV434 cells were maintained in DMEM (Invitrogen, Carlsbad, CA), KGN cells in DMEM/F12 (Invitrogen), and the human umbilical vein endothelial cell line (HUVEC) in M199, all supplemented with 10% fetal bovine serum (Sigma-Aldrich, St Louis, MO), 5% penicillin/streptomycin (Invitrogen), and 1% L-Glutamine (Sigma-Aldrich). HUVEC media was also supplemented with 10 ng/mL fibroblast growth factor and 10 ng/mL epidermal growth factor.
Cell Viability, Proliferation, and Apoptosis Assays
Cells were seeded in 96-well plates at approximately 80% confluency in the appropriate complete medium and incubated in fresh medium containing either the appropriate vehicle control or the specific TKI for the specified length of time.Cellproliferation,cellviability,andapoptosisweremeasured

as described previously.14 Chemical inhibitors used were as follows: sunitinib (Sutent, SU11248; Pfizer, New York, NY), sorafenib(Nexavar,BAY43-9006;Bayer,Leverkusen,Germany), RAF1 kinase inhibitor I (Merck, Darmstadt, Germany), and SU6656 (Src family kinase inhibitor; Merck).
Luciferase Reporter Assays
Cells were transfected with an NFJB-inducible reporter plasmid (pNFJB-Luc; Clontech, Mountain View, CA) and treated with vehicle control or a TKI at the appropriate con- centrationfor24hours,aspreviously.14 Giventhelimitationofa ‘‘squelching’’ effect observed upon cotransfection with a con- stitutively expressing Renilla luciferase construct in the GCT- derived cell lines,14 experiments were carried out in the absence of a transfection efficiency control at least thrice with sextu- plicate wells per point. All values represent the mean T SEM.
Statistical Analysis
Statistical analysis was performed using tools within the GraphPad Prism software package (version 5.03; GraphPad SoftwareInc,SanDiego,CA).Meanswerecomparedusinga1-way analysisofvariancefollowedbyTukeyposthoctestfor multiple comparisons. A P value of less than 0.05 was considered sta- tistically significant. The data are presented as means T SEM.

RESULTS
Effect of Sunitinib on COV434, KGN, and HUVEC Cell Proliferation and Cell Viability
Given the off-target effect observed with imatinib and nilotinib,19 and that at higher concentrations TKI will cross- over to other members of the TK family, we first examined the effect of sunitinib on cell proliferation and cell viability. Sunitinib’s primary targets are the split kinase domain family of RTKs, including the vascular endothelial growth factor re- ceptors 1, 2 and 3 (VEGFR-1, -2, -3), PDGFR> and PDGFRA, KIT, FLT3, and CSF1R.25 This molecular profile provides both overlapping and divergent targets of imatinib, with very different binding affinities in some cases.
The effect of sunitinib on GCT-derived cell proliferation and viability was examined in the COV434 and KGN cell lines. Cells were incubated with increasing concentrations of sunitinib (1Y1000 nM [data not shown]; and 1Y1000 KM) or vehicle (DMSO) alone for 24 hours, at which time cell pro- liferation and viability were assessed (Fig. 1). Sunitinib had no effect on cell proliferation or cell viability in the COV434 and KGN cells, even at a concentration 1000-fold higher than that required to cause 50% inhibition of target kinase auto- phosphorylation (Supplementary Table 1, Supplemental Digi- tal Content 1, available at http://links.lww.com/IGC/A296).
Given the lack of response observed in COV434 and KGN cells, the efficacy of sunitinib was verified using a cell line known to be sensitive to sunitinib. HUVEC were incu- bated with increasing concentrations of sunitinib (1Y1000 nM) or vehicle (DMSO) for 24 hours, at which time cell prolifer- ation and viability was assessed (Fig. 1). HUVEC exhibited a dose-dependent decrease in cell proliferation and cell viability with an IC50 of approximately 30 and 10 nM, respectively. These values are equivalent to that of the published IC50 for

sunitinib (Supplementary Table 1, Supplemental Digital Con- tent 1, available at http://links.lww.com/IGC/A296). Moreover, sunitinib has been shown to inhibit the proliferation of a gas- trointestinal stromal tumor cell line (GIST-T1) with an IC50 of approximately 40 nM,26 confirming that sunitinib elicited an effect at an appropriate concentration.

FIGURE 1. Effect of sunitinib on cell viability and cell proliferation in COV434, KGN, and HUVEC cells.
Cell proliferation in the presence of sunitinib was examined by quantifying the formazan product of MTS reduction after 24 hours ()). Cell viability in the presence of sunitinib was examined by measuring ATP present after 24 hours (h). Each point represents the mean T SEM of 4 separate experiments performed in triplicate, relative to the baseline activity detected in those cells treated with vehicle alone.

1226 * 2015 IGCS and ESGO

KGN cells were transiently transfected with the pNFJB-Luc reporter and treated with vehicle or sorafenib (0.1Y100 KM) for 24 hours (Fig. 3). Although a decrease in luciferase ac- tivity was observed at approximately 30 KM in both cell lines, this concentration is greater than 1000-fold higher than that which was required to inhibit cell proliferation and cell via- bility, indicating that the effect observed on NFJB activity was likely to be due to either cell toxicity or an off-target effect.
Effect of RAF1 Kinase-Specific Inhibitor on COV434 and KGN Cell Proliferation and Cell Viability
The distinct yet overlapping molecular profiles of TKIs, the lack of response seen to imatinib, nilotinib, and sunitinib,

FIGURE 2. Effect of sorafenib on cell viability and cell proliferation in COV434 and KGN cells. Cell proliferation ()) and cell viability (h) were examined as described
in Figure 1. Each point represents the mean T SEM of 4 separate experiments performed in triplicate,
relative to the baseline activity detected in those cells treated with vehicle alone.

Effect of Sorafenib on COV434 and KGN Cell Proliferation, Cell Viability, and NFJB Constitutive Activity
Sorafenib was originally developed as a RAF1 kinase inhibitor (IC50 = 6 nM) and was the first compound of its class to be approved for clinical use. Its targets also include BRAF (22 nM), VEGFR-1 (26 nM), VEGFR-2 (90 nM), VEGFR-3 (20 nM), PDGFRA (57 nM), FLT3 (33 nM), and KIT (58 nM)27; it therefore has a distinct yet overlapping molecular profile to imatinib, nilotinib, and sunitinib. COV434 and KGN cells were incubated with increasing concentrations of sorafenib (1-100 nM) or vehicle (DMSO) alone for 24 hours, before cell proliferation and cell viability were assessed (Fig. 2). Sorafenib dose-dependently inhibited cell viability and pro- liferation in COV434 cells with an IC50 of approximately 10 and 100 nM, respectively, and also inhibited cell viability and proliferation in KGN cells with an IC50 of approxi- mately 10 and 30 nM, respectively (Fig. 2). Interestingly, sorafenib had no effect on apoptosis in either cell line (Sup- plementary Figure 1, Supplemental Digital Content 2, available at http://links.lww.com/IGC/A296).
To determine whether the decrease in cell proliferation and cell viability was due to sorafenib acting via inhibition of the constitutively activated NFJB pathway,14 COV434 and

FIGURE 3. Effect of sorafenib on constitutive NFJB activity in COV434 and KGN cells. Cells transiently transfected with 0.5 Kg of pNFJB-Luc (h) or empty vector (g) were treated with either vehicle or sorafenib (0.1Y100 KM). Firefly luciferase activity was measured and expressed as arbitrary units relative to
vehicle-treated cells. The results from 2 separate experiments, each of which was performed in sextuplicate, are shown as mean T SEM. #Not significant; *P G 0.001 when sorafenib was compared to vehicle alone.

The effects of SU6656, on cell growth and viability were determined in the COV434 and KGN cells. Cells were incu- bated with increasing concentrations of SU6656 (1Y1000 nM) or vehicle (DMSO) alone for 24 hours, before cell prolifera- tion and cell viability were assessed (Fig. 5). SU6656 dose- dependently inhibited cell viability in both cell lines with an IC50 of approximately 30 nM. Interestingly, SU6656 seemed to have a limited effect on cell proliferation, with an IC50 of ap- proximately 300 nM in KGN cells and approaching but not reaching 50% inhibition in COV434 cells. SU6656 had a slight effect on increased apoptosis at 1000 nM (Fig. 6).

DISCUSSION
Most small-molecule TKI have been rationally designed to selectively target the catalytic domain of a specific TK of interest; however, a number approved for the treatment of can- cers are potent inhibitors of multiple kinases. In the clinical setting, these multiple-target activities have proven valuable as they enable a single drug to be used in the treatment of multiple malignancies that may be driven by the aberrant activation of different target kinases.22,28
The aim of this study was to systematically examine the effect of TKI that have overlapping yet divergent target profiles

FIGURE 4. Effect of RAF1 kinase inhibition on cell viability and cell proliferation in COV434 and KGN cells. Cell proliferation ()) and cell viability (h) were examined as described in Figure 1. Each point represents the mean T SEM of 4 separate experiments performed in triplicate, relative to the baseline activity detected in those cells treated with vehicle alone.

together with the dose-dependent decrease in cell proliferation and viability in the nanomolar range observed upon treatment with sorafenib, implicates a role for either RAF1 or BRAF.
Given that abundant RAF1 mRNA expression levels have been observed in both COV434 and KGN cell lines,14 cells were incubated with increasing concentrations of the RAF1-specific inhibitor (1Y1000 nM [data not shown]; and 1Y1000 KM) or vehicle (DMSO) alone for 24 hours, before cell proliferation and cell viability were assessed (Fig. 4). The RAF1 inhibitor had no effect on cell proliferation or viability in the COV434 and KGN cell lines (expected IC50 = 9 nM). The RAF1 inhibitor also had no effect on NFJB constitutive activity (data not shown).

Effect of Src-Family KinaseYSpecific Inhibitor on COV434 and KGN Cell Proliferation and Cell Viability
Having excluded RAF1, we sought to examine the effect of a selective Src-family kinase inhibitor (SU6656) on the COV434 and KGN cell lines. SU6656 is an ATP-competitive, selective inhibitor of Src kinases, including the four ubiquitous- ly expressed Src family members; Src (IC50 = 280 nM), YES (20 nM), LYN (130 nM), and FYN (170 nM), and is a weak inhibitor of LCK (IC50 = 688 KM) and PDGFR (IC50 910 KM).

FIGURE 5. Effect of Src kinase inhibition on cell viability and cell proliferation in COV434 and KGN cells. Cell proliferation ()) and cell viability (h) were examined as described in Figure 1. Each point represents the mean T SEM of 4 separate experiments performed in triplicate, relative to the baseline activity detected in those cells treated with vehicle alone.

1228 * 2015 IGCS and ESGO

FIGURE 6. Effect of Src kinase inhibition on apoptosis in COV434 and KGN cells. Subconfluent COV434 and KGN cells were treated with incremental concentrations of SU6646 for 24 hours 30 minutes before the end of treatment cells were stained with the nuclear dyes, Hoechst 33342 and Yo-Pro-1. Live cell fluorescent microscopy imaging was performed, exciting cells at 340 T 50 nm (for Hoechst 33342) and 455 T 40 nm (Yo-Pro-1) and measuring fluorescence emission at
515 T 20 nm (for both). Cell mortality was quantified by expressing the number of Yo-Pro-1-positive cells as a percentage of the number of Hoechst 33342Ypositive cells (r). Each point represents the mean T SEM of
3 separate experiments performed in triplicate.

on COV434 and KGN cell lines. Given the off-target effect observed with imatinib and nilotinib,19 we first examined the effectofsunitinib(SupplementaryTable1,Supplemental Digital Content 1, available at http://links.lww.com/IGC/A296). Sunitinib had no effect on cell proliferation and cell viability in the COV434 and KGN cell lines. The lack of response observed in COV434 and KGN cells therefore indicates sunitinib’s target TKs are unlikely to be involved in the proliferation of GCT but enables us to narrow the range of potential TK likely to be involved in the pathogenesis of GCT. An important clinical caveat in these studies is that a drug such as sunitinib, which lacks a direct effect on the GCT cell lines, may still have efficacy in vivo, not by directly targeting the tumor cells but through compromising tumor induced neovascularization stimulated by tumor VEGF.

The next TKI examined, sorafenib, dose-dependently inhibited cell proliferation and viability in KGN and COV434 cells. The inhibitory IC50 values are within the range required to elicit 50% inhibition of autophosphorylation of sorafenib’s target TK (Supplementary Table 1, Supplemental Digital Con- tent 1, available at http://links.lww.com/IGC/A296). Sorafenib’s lack of apoptotic effect on in either cell line suggests that its inhibitory action elicits a growth arrest response and not cell death. These data suggest sorafenib’s primary mode of action is via a pro-proliferative pathway rather than a specific ap- optotic pathway, as seen for NFJB inhibition,14 but clearly indicates the potential to exert distinct inhibitory effects in these cell lines.
Given the off-target effect/lack of response observed upon treatment with imatinib, nilotinib, and sunitinib, their primary targets can be ruled out, leaving RAF1 and BRAF as the likely candidates through which sorafenib elicited an in- hibitory response in the COV434 and KGN cell lines. Both kinases can be inhibited by imatinib but only at very high concentrations, consistent with our original observation.16
To further develop this finding, the effect of a RAF1 kinase inhibitor on cell proliferation and cell viability was examined and found to be without effect. These results sug- gest that the inhibitory effect of sorafenib is likely to be via targeted inhibition of BRAF. Chu et al16 also showed that targeting ERK with the chemical inhibitor, PD98059, fully abrogated the constitutive activity of AP-1 in both COV434 and KGN cell lines. Therefore, a proposed pathway to the activation of the AP-1 transcription factor is via RTK/SRC/
RAS/BRAF/MEK/ERK in the GCT cell lines.
Aberrantly activated components of the MAPK path- way are well described in many forms of human cancer; in particular, the mutational ‘‘hotspots’’ include NRAS, KRAS, HRAS, and BRAF. BRAF is the only RAF protein to be frequently mutated in cancer.29Y31 The most common BRAF mutation corresponds to a T 9 A transversion at position 1799, resulting in the substitution of valine by glutamate at position 600 of the protein. However, we did not find the V600E mutation or over expression of BRAF to be a feature of GCT.32 It may be that other known BRAF mutations, of which there are approximately 30, contribute to unopposed activa- tion of the protein, or that aberrant activation originates upstream of BRAF in the MAPK pathway. We sought to examine the effect of a selective Src-family kinase inhibitor (SU6656) on the COV434 and KGN cell lines. Interestingly, the results of the cell proliferation and cell viability assays in the presence of SU6656 somewhat parallel those seen for sorafenib; that is, although SU6656 inhibited cell viability in both cell lines with an IC50 of approximately30 nM, it had limited effect on cellular proliferation and apoptosis. The effect is not as profound as that observed with sorafenib, suggesting that Src is a contributor to the BRAF activity but not the rate-limiting step. Indeed of SU6656’s 4 target TKs, the IC50 argues for a role for Yes.
Given that both NFJB and AP-1 are constitutively activated in COV434 and KGN cells,14,16 and constitutive activation of ERK-1/2 has also been reported in KGN cells,33 sorafenib becomes a powerful tool to aid the characterization of these pathways in the GCT-derived cell lines. In vitro,

sorafenib has been shown to interrupt MEK-1 and ERK-1 phosphorylation in a number of cell lines,34 and given the strong, albeit circumstantial evidence that constitutively activated AP-1 signals via the MAPK/ERK pathway in GCT, the cells’ response to sorafenib may shed light on the mechanism(s) driving the constitutive activity. Although COV434 and KGN cells treated with sorafenib exhibited a decrease in NFJB activity at approximately 30 and 10 KM, respectively, at these concentrations, inhibition of the NFJB pathway is likely to be due to an off-target effect. Therefore, the dose-dependent decrease observed in cell proliferation and cell viability may be due to sorafenib’s target TK sig- naling via the MAPK pathway. These data also suggest that the mechanisms contributing to constitutive NFJB and AP-1 activity are likely to be independent, rather than a single unopposed component eliciting cross-talk between both pathways.
The current findings argue that BRAF inhibition may represent a useful strategy in the treatment of GCT. It will be important to confirm the central role of BRAF inhibition using either siRNA and/or an independent chemical inhibitor of BRAF. However, identification of the basis of the activation of this signaling pathway as a driver of cell proliferation is more complex due to a number of potential points of inter- action. The effects observed with the Src inhibitor suggest that the aberrant activation occurs upstream of BRAF and RAS in the pathway, either in the Src complex or perhaps in an as- sociated receptor TK. Thus, knockdown of RAS may present another option, however, other pathways that may also signal through RAF should be considered, including, for instance, the TAK-TAB complex.
Several case reports indicate that the angiogenesis in- hibitor bevacizumab, a monoclonal antibody against VEGF, has been used successfully in the treatment of refractory GCT.35Y37 Furthermore, a detailed in vitro study identified an autocrine role for VEGF in GCT.38 Although bevacizumab blocks the actions of VEGF, interestingly, we did not observe an effect of sunitinib which acts at the level of the VEGFR TK in the cells. Taken together, these data suggest that the bevacizumab effect does not involve VEGFR. It is also cu- rious that the autocrine stimulation seems to be primarily antiapoptotic, given that the primary effect of bevacizumab was on apoptosis.38 We have demonstrated that the GCT- derived cell lines also have a constitutive NFJB signaling and that when this pathway is blocked, marked apoptosis is seen.14 This raises the possibility that the signaling blocked by bevacizumab is acting via the NFJB pathway rather than the conventional VEGF RTK.
In conclusion, this study suggests that small-molecule TKI already in clinical use may be an effective therapeutic in the treatment of GCTand also provides further insights into the mechanism of aberrant NFJB and AP-1 pathway acti- vation observed in COV434 and KGN cell lines.

ACKNOWLEDGMENTS
The authors thank Dr Simon Chu for the helpful dis- cussions and Maria Alexiadis for the outstanding technical assistance.

REFERENCES
1.Chu S, Rushdi S, Zumpe ET, et al. FSH-regulated gene expression profiles in ovarian tumours and normal ovaries. Mol Hum Reprod. 2002;8:426Y433.
2.Fuller PJ, Chu S, Fikret S, et al. Molecular pathogenesis of granulosa cell tumours. Mol Cell Endocrinol. 2002;191:89Y96.
3.Al-Agha OM, Huwait HF, Chow C, et al. FOXL2 is a sensitive and specific marker for sex cord-stromal tumors of the ovary. Am J Surg Pathol. 2011;35:484Y494.
4.Gershon R, Aviel-Ronen S, Korach J, et al. FOXL2 C402G mutation detection using MALDI-TOF-MS in DNA extracted from Israeli granulosa cell tumors. Gynecol Oncol. 2011;122:580Y584.
5.Hes O, Vanecek T, Petersson F, et al. Mutational analysis (c.402C9G) of the FOXL2 gene and immunohistochemical expression of the FOXL2 protein in testicular adult type granulosa cell tumors and incompletely differentiated sex cord stromal tumors. Appl Immunohistochem Mol Morphol. 2011;19:347Y351.
6.Jamieson S, Butzow R, Andersson N, et al. The FOXL2 C134W mutation is characteristic of adult granulosa cell tumors of
the ovary. Mod Pathol. 2010;23:1477Y1485.
7.Kim MS, Hur SY, Yoo NJ, et al. Mutational analysis of FOXL2 codon 134 in granulosa cell tumour of ovary and other human cancers. J Pathol. 2010;221:147Y152.
8.Kim T, Sung CO, Song SY, et al. FOXL2 mutation in granulosa-cell tumours of the ovary. Histopathology. 2010;56:408Y410.
9.Schrader KA, Gorbatcheva B, Senz J, et al. The specificity of the FOXL2 c.402C9G somatic mutation: a survey of solid tumors. PLoS One. 2009;4:e7988.
10.Shah SP, Kobel M, Senz J, et al. Mutation of FOXL2 in granulosa-cell tumors of the ovary. N Engl J Med. 2009;360:2719Y2729.
11.Oparka R, Cassidy A, Reilly S, et al. The C134W (402 C9G) FOXL2 mutation is absent in ovarian gynandroblastoma: insights into the genesis of an unusual tumour. Histopathology. 2012;60:838Y842.
12.Geiersbach KB, Jarboe EA, Jahromi MS, et al. FOXL2 mutation and large-scale genomic imbalances in adult granulosa cell tumors of the ovary. Cancer Genet. 2011;204:596Y602.
13.Jamieson S, Fuller PJ. Molecular pathogenesis of granulosa cell tumors of the ovary. Endocr Rev. 2012;33:109Y144.
14.Jamieson S, Fuller PJ. Characterization of the inhibitor of kappaB kinase (IKK) complex in granulosa cell tumors of the ovary and granulosa cell tumor-derived cell lines. Horm Cancer. 2013;4:277Y292.
15.Benayoun BA, Caburet S, Dipietromaria A, et al. Functional exploration of the adult ovarian granulosa cell tumor-associated somatic FOXL2 mutation p.Cys134Trp (c.402C9G). PLoS One. 2010;5:e8789.
16.Chu S, Nishi Y, Yanase T, et al. Transrepression of estrogen receptor beta signaling by nuclear factor-JB in ovarian granulosa cells. Mol Endocrinol. 2004;18:1919Y1928.
17.Jakob A, Geiger R, Freidrich WH. Successful treatment of a patient with a granulosa/theca cell tumor of the ovary with STI571 (Gleevic). Proc Am Soc Clin Oncol. 2002;21:24b.
18.Raspagliesi F, Martinelli F, Grijuela B, et al. Third-line chemotherapy with tyrosine kinase inhibitor (imatinib mesylate) in recurrent ovarian granulosa cell tumor: case report.
J Obstet Gynaecol Res. 2011;37:1864Y1867.
19.Chu S, Alexiadis M, Fuller PJ. Expression, mutational analysis and in vitro response of imatinib mesylate and nilotinib target genes in ovarian granulosa cell tumors. Gynecol Oncol. 2008;108:182Y190.

1230 * 2015 IGCS and ESGO

20.Capdeville R, Buchdunger E, Zimmermann J, et al. Glivec (STI571, imatinib), a rationally developed, targeted anticancer drug. Nat Rev Drug Discov. 2002;1:493Y502.
21.Keshava N, Gubba S, Tekmal RR. Overexpression of macrophage colony-stimulating factor (CSF-1) and its receptor, c-fms, in normal ovarian granulosa cells leads to cell proliferation and tumorigenesis. J Soc Gynecol Investig. 1999;6:41Y49.
22.Ja¨nne PA, Gray N, Settleman J. Factors underlying sensitivity of cancers to small-molecule kinase inhibitors. Nat Rev Drug Discov. 2009;8:709Y723.
23.van den Berg-Bakker CA, Hagemeijer A, Franken-Postma EM, et al. Establishment and characterization of 7 ovarian carcinoma cell lines and one granulosa tumor cell line: growth features and cytogenetics. Int J Cancer. 1993;53:613Y620.
24.Nishi Y, Yanase T, Mu Y, et al. Establishment and characterization of a steroidogenic human granulosa-like tumor cell line, KGN, that expresses functional follicle-stimulating hormone receptor. Endocrinology. 2001;142:437Y445.
25.Chow LQ, Eckhardt SG. Sunitinib: from rational design to clinical efficacy. J Clin Oncol. 2007;25:884Y896.
26.Ikezoe T, Yang Y, Nishioka C, et al. Effect of SU11248 on gastrointestinal stromal tumor-T1 cells: enhancement of growth inhibition via inhibition of 3-kinase/Akt/mammalian target of rapamycin signaling. Cancer Sci. 2006;97:945Y951.
27.Stein MN, Flaherty KT. CCR drug updates: sorafenib and sunitinib in renal cell carcinoma. Clin Cancer Res. 2007;13:3765Y3770.
28.Arora A, Scholar EM. Role of tyrosine kinase inhibitors
in cancer therapy. J Pharmacol Exp Ther. 2005;315:971Y979.
29.Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature. 2002;417:949Y954.
30.Cohen Y, Xing M, Mambo E, et al. BRAF mutation in papillary thyroid carcinoma. J Natl Cancer Inst. 2003;95:625Y627.

31.Kimura ET, Nikiforova MN, Zhu Z, et al. High prevalence of BRAF mutations in thyroid cancer: genetic evidence for constitutive activation of the RET/PTC-RAS-BRAF signaling pathway in papillary thyroid carcinoma. Cancer Res. 2003;63:1454Y1457.
32.Jamieson S, Alexiadis M, Fuller PJ. Expression status and mutational analysis of the ras and B-raf genes in ovarian granulosa cell and epithelial tumors. Gynecol Oncol. 2004;95:603Y609.
33.Steinmetz R, Wagoner HA, Zeng P, et al. Mechanisms regulating the constitutive activation of the extracellular signal-regulated kinase (ERK) signaling pathway in ovarian cancer and the effect of ribonucleic acid interference for ERK1/2 on cancer cell proliferation. Mol Endocrinol. 2004;18:2570Y2582.
34.Wilhelm SM, Carter C, Tang L, et al. BAY 43-9006 exhibits broad spectrum oral antitumor activity and targets the RAF/
MEK/ERK pathway and receptor tyrosine kinases involved in tumor progression and angiogenesis. Cancer Res. 2004;64:7099Y7109.
35.Barrena Medel NI, Herzog TJ, Wright JD, et al. Neoadjuvant bevacizumab in a granulosa cell tumor of the ovary: a case report. Anticancer Res. 2010;30:4767Y4768.
36.Kesterson JP, Mhawech-Fauceglia P, Lele S. The use of bevacizumab in refractory ovarian granulosa-cell carcinoma with symptomatic relief of ascites: a case report. Gynecol Oncol. 2008;111:527Y529.
37.Tao X, Sood AK, Deavers MT, et al. Anti-angiogenesis therapy with bevacizumab for patients with ovarian granulosa cell tumors. Gynecol Oncol. 2009;114:431Y436.
38.Farkkila A, Pihlajoki M, Tauriala H, et al. Serum vascular endothelial growth factor A (VEGF) is elevated in patients with ovarian granulosa cell tumor (GCT), and VEGF inhibition by bevacizumab induces apoptosis in GCT in vitro. J Clin Endocrinol Metab. 2011;96:E1973YE1981.