Oncogenic PIK3CA gene mutations and HER2/neu gene amplifications determine the sensitivity of uterine serous carcinoma cell lines to GDC-0980, a selective inhibitor
of Class I PI3 kinase and mTOR kinase (TORC1/2)
Diana P. English, MD; Stefania Bellone, PhD; Emiliano Cocco, PhD; Ileana Bortolomai, PhD; Sergio Pecorelli, MD; Salvatore Lopez, MD; Dan-Arin Silasi, MD; Peter E. Schwartz, MD; Thomas Rutherford, MD; Alessandro D. Santin, MD

OBJECTIVE: To evaluate PIK3CA mutational status and c-erbB2 gene amplification in a series of primary uterine serous carcinomas (USC) cell lines. To assess the efficacy of GDC-0980, a potent inhibitor of Class I PI3 kinase and mTOR kinase (TORC1/2), against primary USC harboring HER2/neu gene amplification and/or PIK3CA mutations.
STUDY DESIGN: Twenty-two primary USC cell lines were evaluated for c-erbB2 oncogene amplification by fluorescence in situ hybridization (FISH) assays and for PIK3CA gene mutations by direct DNA sequencing of exons 9 and 20. In vitro sensitivity to GDC-0980 was evaluated by flow-cytometry-based viability and proliferation assays. Downstream cellular responses to GDC-0980 were assessed by measuring phosphorylation of the 4-EBP1 protein by flow-cytometry.
RESULTS: Five of 22 (22.7%) USC cell lines contained oncogenic PIK3CA mutations although 9 (40.9%) harbored c-erbB2 gene
amplification by FISH. GDC-0980 caused a strong differential growth inhibition in FISHþ USC when compared with FISHti (GDC- 0980 IC50 mean ti SEM ¼ 0.29 ti 0.05 mM in FISHþ vs 1.09 ti 0.20 mM in FISHti tumors, P ¼ .02). FISHþ USC harboring PIK3CA mutations were significantly more sensitive to GDC-0980 exposure when compared with USC cell lines harboring wild-type PIK3CA (P ¼ .03). GDC-0980 growth-inhibition was associated with a significant and dose-dependent decline in phosphorylated 4-EBP1 levels.
CONCLUSION: Oncogenic PIK3CA mutations and c-erbB2 gene amplification may represent biomarkers to identify patients harboring USC who may benefit most from the use of GDC-0980.
Key words: endometrial neoplasms, uterine serous tumors, HER2/
neu, PIK3CA, mTOR inhibitor, GDC-0980

Cite this article as: English DP, Bellone S, Cocco E, et al. Oncogenic PIK3CA gene mutations and HER2/neu gene amplifications determine the sensitivity of uterine serous carcinoma cell lines to GDC-0980, a selective inhibitor of Class I PI3 kinase and mTOR kinase (TORC1/2). Am J Obstet Gynecol 2013;209:465.e1-9.

ndometrial carcinoma is one of the most common gynecologic cancers.
About 49,560 new cases of endometrial cancer will be diagnosed in the year 2013 in the United States, with approximately 8190 deaths.1 Type I endometrial cancers account for most of these tumors
and, typically presents as well or moderately differentiated carcinomas; these tumors are endometrioid in his- tology, are associated with endogenous hyperestrogenic conditions or chronic exposure to unopposed estrogens and are generally highly curable with
primary surgery.2 In contrast, type II endometrial cancers account for a mi- nority of cases and are not associated with unopposed estrogen exposure; these tumors have often an unfavorable prognosis because of their aggressive biologic behavior and consequent high risk for recurrence.2 Uterine serous car- cinoma (USC) represents the most

From the Department of Obstetrics, Gynecology, and Reproductive Sciences, Yale University School of Medicine, New Haven, CT (Drs English, Bellone, Cocco, Bortolomai, Silasi, Schwartz, Rutherford, and Santin), and the Division of Gynecologic Oncology, University of Brescia, Brescia (Dr Pecorelli), and the Division of Gynecologic Oncology, University Campus Biomedico of Rome, Rome (Dr Lopez), Italy.
Received May 8, 2013; revised July 6, 2013; accepted July 22, 2013. The authors report no confl ict of interest.
This work was supported in part by grant numbers NIH R01 CA122728-01A4 and R01 CA154460-01A1 (ADS) and NIH Research Grant CA-16359 from the NCI. Genentech-Roche provided GDC-0980 gratis.
Presented in oral format at the 34th Annual Scientifi c Meeting of the New England Association of Gynecologic Oncologists, Westbrook, CT, June 14-16, 2013.
Reprints are not available from the authors.
0002-9378/$36.00 ti ª 2013 Mosby, Inc. All rights reserved. ti http://dx.doi.org/10.1016/j.ajog.2013.07.020
aggressive variant of type II endometrial cancer.3 Although this tumor subtype constitutes less than 10% of all endo- metrial tumors, it is responsible for 40% of deaths because of endometrial cancer.4,5 The discovery of molecular signaling pathways highly active in uter- ine serous tumors may lead to the devel- opment of novel, specific, and more effective therapeutic strategies against this aggressive subset of uterine carcinoma.
The phosphatidylinositol 3-kinase/
AKT (PIK3/AKT)-mammalian target of rapamycin (mTOR) signaling cascade

plays a central role in diverse cellular responses such as proliferation, survival, mobility, metabolism, and control of malignant cellular growth.6,7 HER2/neu, a member of the erbB receptor tyrosine kinase (TK) family located upstream to the PIK3CA/AKT/mTOR pathway, is associated with cancer cell proliferation, poor survival, and resistance to therapy in multiple human tumors.8 HER2/neu and the PIK3CA/AKT/mTOR pathway are often constitutively activated in various human cancers secondary to gene amplifi cations (ie, HER2/neu) or activating mutations in the PIK3CA/
AKT genes, providing novel targets for cancer therapy.9,10
Importantly, multiple research groups including our own have recently re- ported presence of PIK3CA gene muta- tions and HER2/neu gene amplifi cations in a relevant number of USC by whole exome sequencing.11-13 These data pro- vide evidence to suggest that the use of PIK3/AKT/mTOR inhibitors may pro- vide anticancer activity in biologically aggressive Type II endometrial cancers such as USC.
GDC-0980 is a novel, potent, orally available small molecule inhibitor with selectivity for class I PIK3CA and mTOR kinase. GDC-0980 has shown remark- able ability to target a variety of human cancer cell lines through the inhibition of both PIK3CA and mTOR signaling.14,15 GDC-0980 strongly inhibited the PI3K pathways and the phosphorylation of mTORC1 substrates S6K and 4E-BP1 as well as phosphorylation of the mTORC2 substrate AKT and down- stream proteins.14-15 In vivo, GDC-0980 has demonstrated potent antitumor efficacy in nude mice bearing multiple human xenografts of breast cancer, nonsmall cell lung cancer, pancreatic cancer, colon cancer, prostate cancer, and melanoma.14,15 On the basis of these promising preclinical results, GDC-0980 is currently being tested in phase II studies in patients with advanced solid tumors.
Although multiple mTOR and/or PIK3CA inhibitors are currently under evaluation in clinical trials against a variety of human cancers including endometrial cancer, to our knowledge,

no study has yet analyzed and compared the in vitro sensitivity of primary USC cell lines to selective dual PIK3CA/
mTOR inhibitors. More importantly, no study has yet evaluated whether bio- logically aggressive USC harboring PIK3CA mutations and/or HER2/neu gene amplifi cations are selectively sen- sitive to agents targeting the PIK3CA/
mTOR pathway. To fi ll this gap in knowledge, in this study we have: (1) sequenced a large panel of primary USC cell lines for oncogenic PIK3CA muta- tions, (2) evaluated c-erbB2 oncogene amplifi cation by FISH assays, (3) eval- uated and compared the sensitivity to GDC-0980 in USC harboring amplifi – cation of c-erbB2 and/or PIK3CA mu- tations vs USC control cell lines, and (4) analyzed the baseline levels and changes in phosphoprotein 4E-BP1 expression as a downstream cellular response to GDC-0980 exposure in vitro. We report the fi rst evidence that oncogenic PIK3CA gene mutations and HER2/neu gene amplifi cations may be predictive of the sensitivity of USC cell lines to GDC-0980. Class I PIK3CA/mTOR ki- nase inhibitors such as GDC-0980 may represent promising novel drugs in pa- tients harboring biologically aggressive USC with PIK3CA mutations and amplifi cation of c-erbB2.

MATERIALS AND METHODS Establishment of USC cell lines and evaluation for HER2/neu gene amplification
Twenty-two primary uterine papillary serous carcinoma cell lines were evalu- ated in our study. Specimens were ob- tained from fresh tumor biopsies collected at the time of surgery, under approval of the institutional review board. The cells were maintained as a monolayer in RPMI-1640 medium (Mediatech, Manassas, VA) supple- mented with 10% fetal bovine serum (Gemini, Woodland, CA), 1% antibiotic (penicillin, streptomycin), and 0.3% antimycotic amphotericin B (Invitrogen, Carlsbad, CA). Cells were incubated at 37ti C in a humidifi ed atmosphere of 95% air/5% CO2. Source-patient character- istics of the USC cell lines are described in the Table. Tumor c-erbB2 gene

amplification was evaluated by FISH as previously described by our group16 in all cell lines and results are presented in the Table.

Polymerase chain amplification Primers for polymerase chain reaction (PCR) amplification and sequencing were designed using the Primer3 program (http://frodo.wi.mit.edu/cgi-bin/primer3/
primer3_www.cgi) and were synthesized by the Yale Keck facility (http://keck.med. yale.edu/), based on PIK3CA sequence obtained from National Center for Bio- technology Information (accession num- ber N M_006218). Two primer pairs were used to individually amplify exon 9 and exon 20 of PIK3CA from genomic USC primary cell lines DNA (Table). PCR amplification was performed in 20 mL reaction volumes that contained 100 ng of DNA, 75 mmol/L Tris-HCl, 1.5 mmol/L MgCl2, 50 mmol/L KCl, 20 mmol/L (NH4)2SO4, 0.2 mmol/L of each primer, 0.2 mmol/L of each deoxyribonucleotide triphosphate, and 1 U of Taq DNA poly- merase (BIOTOOLS; B&M Laboratories, SA, Madrid, Spain). The 2 exons were amplified with the following PCR con- ditions: an initial 5-minute denaturation at 94ti C followed by 35 cycles of 1 minute at 94ti C, 1 minute at 57ti C, 1 minute at 72ti C, and a final extension of 10 minutes at 72ti C.

DNA sequencing
PCR products were first purified using the MinElute PCR Purification Kit (Qiagen GmbH, Hilden, Germany) and were bidirectionally sequenced using the orig- inal primer pair and Applied Biosystem Cycle Sequencing kit (Applied Biosystem Inc, Santa Clara, CA) at the Yale Keck fa- cility (http://keck.med.yale.edu). Samples were analyzed on the ABI Prism 3100 Avant instrument (Applied Biosystems), using standard run parameters. The sep- aration matrix used was POP-6 using 1ti Tris-borate-EDTA buffer with EDTA running buffer (Applied Biosystems).

Fluorescent in situ hybridization
FISH analysis was performed using the PathVysion HER2 DNA FISH Kit (Abbott Molecular Inc., Abbott Park, IL) according to the manufacturer’s

instructions. Fluorescent signals in at

least 30 nonoverlapping interphase nuclei with intact morphology were scored. Tumor cells were scored for the number of orange (HER2) and green (chromosome 17) signals. A case was
TABLE
Patient characteristics, c-erbB2 gene amplification and PIK3CA mutations in primary USC cell lines
Sample ID Age Race FIGO stage PIK3CA mutations Exon 9,20 C-erbB2 FISH
USPC-ARK1a 62 B IV 542/1068 Amplified

scored as amplified when the ratio of the number of fluorescent signals of HER2
USPC-ARK2a 63 B
IV
Not detected
Amplified

to chromosome 17 was ti 2.
USPC-ARK3a 59 B IV
USPC-ARK4a 73 W IV
Not detected Not detected
Amplified Not amplified

Drug
USPC-ARK5a 73 B IIIC Not detected Not amplified

GDC-0980 was provided by Genentech Inc (South San Francisco, CA). GDC- 0980 was dissolved in dimethyl sulf- oxide (Sigma-Aldrich, St. Louis, MO) as a 10-mM stock solution and was diluted in culture medium immediately before addition to cell lines.
USPC-ARK6a 62 W IB
USPC-ARK7a 75 W IIC
USPC-ARK8a 88 W IIIA
USPC-ARK9 73 B IIC
USPC-ARK10 79 W IVB
Not detected Not detected Not detected 1044/1068 1044/1068
Not amplified Not amplified Not amplified Amplified Amplified

USPC-ARK11a 80 B IIIC Not detected Not amplified

Primary USC growth rate analysis
USPC-ARK12 80 B IIIC Not detected Not amplified

The doubling time at the log phase of growth for the c-erbB2 FISH þ USC cell lines was determined by counting the number of live cells at different time points after plating. Briefly, cell lines were cultured in RPMI 1640 (Mediatech, Manassas, VA) supplemented with 10% fetal bovine serum (Gemini, Woodland, CA). When cultures had grown to approximately 80% confl uence, cells were harvested from the fl ask using 0.25% Trypsin EDTA (Invitrogen), then counted on a hemocytometer chamber
USPC-ARK13 67 W IVB
USPC-ARK14a 73 B IV
USPC-ARK15a 67 W IIIC
USPC-ARK16 65 W IV
USPC-ARK17 59 W IIIA
USPC-ARK18 62 B IIB
USPC-ARK19a 65 W IA
USPC-ARK20a 42 W II
USPC-ARK21a 70 W IA
USPC-ARK22a 60 W IVB
Not detected 546/1068 Not detected Not detected Not detected Not detected Not detected 1047/1068 Not detected Not detected
Not amplified Not amplified Not amplified Not amplified Amplified Amplified
Not amplified Amplified Amplified
Not amplified

and assessed for viability via Trypan blue exclusion. Cell density was adjusted to a concentration of 10,000 cells/mL, then cells were seeded into a 6-well mL plate (Corning Life Sciences, Lowell, MA) at a density of 40,000 cells per well.
B, black; FIGO, International Federation of Gynecology and Obstetrics stage 1988; FISH, fluorescence in situ hybridization; USC, uterine serous carcinomas; USPC, uterine serous papillary carcinoma; W, white.
a USC cell lines established as long-term cultures used in chemosensitivity experiments with GDC-0980.
English. GDC-0980, a PIK3CA/mTOR inhibitor, is highly active against USC in vitro. Am J Obstet Gynecol 2013.

Three replicate wells were plated per cell line per time point. Cell plates were incubated at 37ti C with 5% CO2 for 24, 48, 72, 96, and 120 hours, at which time they were removed from the incubator, harvested from the wells using 0.25% Trypsin EDTA, counted on a hemocytometer chamber and assessed for viability via Trypan blue exclusion.

Chemo-response assay
The effect of GDC-0980 on the viability and IC50 of cells was determined using fl ow cytometry assays as previously described.17 Briefl y, tumor cells derived from 15 primary USC cell lines
established as long term cultures in vitro (ie, 5 cell lines harboring HER2/neu gene amplifi cations vs 10 HER2/neu-negative cell lines) were plated in 6-well tissue culture plates and when in exponential growth treated with GDC-0980 at con- centrations of 0.01 0.25, 0.5, 0.75, 1.0, 2.0 mM. Concentrations up to 4.0 mM were also used against the more resistant USC cell lines. After 72 hours of addi- tional incubation, well contents were harvested in their entirety, centrifuged then stained with propidium iodide (2 mL of a 500 mg/mL stock solution in PBS with 0.1% sodium azide and 2% fetal bovine serum) for flow cytometric counts. Viable cells were then quantified
using flow cytometry as percent of viable cells (mean ti SEM) after exposure to different concentrations of GDC-0980 relative to vehicle-treated cells taken as 100% viable. A minimum of 3 inde- pendent experiments per USC cell line were performed.
Flow cytometry analysis of phosphorylated 4-EBP1 intracellular levels in primary USC cell lines
A previously validated flow cytometry- based assay18 was used to evaluate the baseline level as well as the change in phosphorylated 4-EBP1 expression as a downstream cellular response to GDC- 0980 in USC cell lines. Briefly, USC

cells after 72 hours exposure to 0.5 mM,

FIGURE 1
Inhibition of uterine serous carcinoma cell lines proliferation after exposure to GDC-0980

Graph showing IC50 values for all 15 USC cell lines following exposure to GDC-0980. Inhibition of USC cell proliferation was determined by flow cytometry. The 5 c-erbB2 FISHþ cell lines (USC ARK- 1, USC ARK-2, USC ARK-3, USC ARK-20, and USC ARK-21) are located to the left of the graph and the 10 c-erbB2 negative USC cell lines are located to the right of the graph. The asterisk denotes the most sensitive and most resistant cells lines in the c-erbB2 FISHþ and FISHti groups, respectively. Cells were treated with GDC-0980 and incubated for 72 hrs. Each column on the graph represents
the mean of 3 experiments in each group. Mean IC50 values of c-erbB2 FISHþ compared with c-erbB2 FISHti cell lines are statistically significantly different (P ¼ .02).
FISH, fluorescence in situ hybridization; USC, uterine serous carcinoma.
English. GDC-0980, a PIK3CA/mTOR inhibitor, is highly active against USC in vitro. Am J Obstet Gynecol 2013.
1.0 mM, and 2.0 mM of GDC-0980 were fixed in 4% formaldehyde and per- meabilized with ice-cold 100% methanol. GDC-0980 treated and untreated control cells were incubated with primary rabbit monoclonal antibody against 4-EBP1 (Cell Signaling Technology, Inc., Danvers, MA) following the protocol provided by the manufacturers and stained with a fluorescein isothiocyanate-conjugated goat antirabbit F(ab’)2 immunoglobulin as a secondary reagent (Chemicon Inter- national, Temecula, CA). Cells (ie, 10,000 events per sample) were analyzed on FACSCalibur, using Cell Quest software (BD Biosciences, San Jose, CA).

Statistical analysis
Statistical analysis was conducted using GraphPad Prism5 (GraphPad Software, Inc., San Diego, CA). For each indepen- dent experiment of GDC-0980 on a given cell line, the measures of growth under different dose levels were normalized to the mean of the control group receiving no drug, so that all data were expressed as a proportion of the control. Normalized data then were fit via nonlinear regres- sion to a 3-parameter logistic response curve against the base-10 logarithms of dose in micrometers, and the resulting parameter estimates were used to calcu- late the value of the IC50 (in log10 units) for that experiment. The collection of log10 (IC50)s thereby determined were compared between each cell line. For the flow cytometry experiments, changes in the phosphorylated protein 4-EBP1 levels were analyzed comparing the mean intensity of fluorecence (MFI) before and after the exposure with GDC-0980. Unpaired t test was used to compare 4-EBP1 changes in cell lines exposed with GDC-0980. Differences in all compari- sons were considered significant at P values < .05.

RESULTS
Molecular results
Mutations in the PIK3CA gene were observed in 5 primary USC cell lines (22.7%) (Table). Three mutations were located in exon 20 and 2 in exon 9

(Table). All were missense mutations.

The 1044/1068 mutation was identifi ed twice. The remaining tumors contained the 542/1068, the 546/1068, and 1047/
1068 mutations. c-erbB2 gene amplifi - cation by FISH was identified in 9 of 21 (40.9%) primary USC cell lines. Of the c-erbB2 amplified tumors, 4 of 9 (44.4%) were found to harbor PIK3CA muta- tions in exon 9 or 20. In contrast, only 1 of the 13 (7.6%) USC cell lines showing no c-erbB2 amplification was found to harbor mutations in exon 9 or exon 20.

Growth curve
Five of 9 short-term USC primary cell lines showing c-erbB2 amplification were successfully established as long-term cultures and used in the in vitro experi- ments with GDC-0980 described below. Two of 5 (ie, uterine serous papillary carcinoma [USPC] ARK-1 and USPC ARK-20) harbored PIK3CA gene muta- tions (Table). The doubling time of all 5 primary USC cell lines was calculated at the log phase of growth by counting the number of live cells 24, 48, 72, 96, and 120 hours after plating as described in the Methods section. Although all these tumors demonstrated fast growth rates, population doubling times were similar: USPC ARK-1 ¼ 18.5 hours, USPC ARK- 2 ¼ 19.3 hours, USPC ARK-3 ¼ 18.3 hours, USPC ARK-20 ¼ 17.8 hours, and USPC-ARK-21 ¼ 17.4 hours (P ¼ not signifi cant).
HER2/neu gene amplification and PIK3CA gene mutations determine response to GDC-0980
FIGURE 2
Dose response curves of uterine serous carcinoma cell lines after exposure to GDC-0980

A, Representative dose response curves with IC50 of a PI3KCA mutated cell line (ie, USC ARK-20, upper left panel ) vs a PI3KCA wild-type cell line (ie, USC ARK-2, upper right panel ). B, Scatter
ICplot showing IC50 (mean ti SEM) for PI3KCA mutated c-erbB2 amplified cell lines compared with the
50
are c-erbB2 amplified and demonstrated similar growth rates, the PI3KCA mutated cell lines are significantly more sensitive than PI3KCA wild-type cell lines with GDC-0980 (P ¼ .03).
English. GDC-0980, a PIK3CA/mTOR inhibitor, is highly active against USC in vitro. Am J Obstet Gynecol 2013.

We initially compared the effects of scalar concentrations of GDC-0980 in all 15 primary USC celllines established as long- term in vitro cultures (Table). As repre- sentatively shown in Figure 1, we found GDC-0980 to cause a significantly stron- ger differential growth inhibition in
c-erbB2 FISHþ USC cell lines when
50
mean ti SEM ¼ 0.29 ti 0.05 mM in FISHþ vs 1.09 ti 0.20 mM in FISHti tu- mors, P ¼ .02). The c-erbB2 FISHþ USC cell lines, as a group, were therefore more than 3 times more sensitive to GDC- 0980 than the c-erbB2 FISHe cell lines (Figure 1, lower panel). USPC-ARK-1
cell line was the most sensitive to GDC- 0980, with a mean inhibitory concen- tration (IC50) ti standard error (SEM) of 0.14 ti 0.006 mM followed by the ARK- 20 cell line (mean IC50 value of 0.22 ti 0.005 mM). USPC-ARK-5 and USPC- ARK-8 cell lines were found to be the least sensitive, with a mean IC50 of 2.09 ti 0.25 and 3.12 ti 0.30 mM, respectively (Figure 1, upper panel).
We next investigated whether FISHþ USC harboring PIK3CA mutations were significantly more sensitive to GDC- 0980 exposure when compared with FISHþ USC cell lines harboring wild- type PIK3CA. In multiple experiments,
as a group, we found c-erbB2 amplifi ed cell lines harboring PIK3CA mutations (ie, USPC-ARK-1, USPC-ARK-20) to be signifi cantly more sensitive to GDC- 0980 (ie, mean inhibitory concentra- tion [IC50] ti SEM ¼ 0.18 ti 0.04 mM) when compared with the PIK3CA wild-type c-erbB2 amplified cell lines USPC-ARK-2, USPC-ARK-3, and USPC- ARK21 (ie, mean IC50 values of 0.37 ti 0.03 mM, (P ¼ .03) (Figure 2).
None of our PI3K mutated cell lines had also a KRAS mutation and KRAS mutational status did not seem to inde- pendently affect the sensitivity to GDC-0980.

FIGURE 3
Decrease in p4-EBP1 in uterine serous carcinoma cells lines after exposure to GDC-0980

Graphs showing a statistically significant difference in the reduction in mean fluorescence intensity (MFI) of A, p4-EBP1 in PI3KCA mutated cells lines and B, PI3KCA wild-type cell lines. C, The decrease in MFI of p4-EBP1 after exposure to GDC-0980 compared with untreated controls for all
cell lines is shown (P ¼ .01).
English. GDC-0980, a PIK3CA/mTOR inhibitor, is highly active against USC in vitro. Am J Obstet Gynecol 2013.

GDC-0980 affects p4E-BP1 expression levels in USC cell lines
We next evaluated p4E-BP1 expression levels by flow cytometry in all c-erbB2 amplified USC cell lines established as long-term cultures harboring PIK3CA mutations or wild-type PIK3CA before exposure to GDC-0980. No significant differences were found in the level of MFI of the phosphorylated protein 4E-BP1 among the group of cell lines harboring PIK3CA mutations when compared with those harboring wild-type PIK3CA (data not shown). Next, we evaluated the phosphorylation of 4E-BP1 after expo- sure to increasing concentrations of GDC-0980 (0.5, 1.0, and 2.0 mM). Regardless of the presence or absence of the PIK3CA mutation, GDC-0980 was able to reduce 4E-BP1 phosphorylation in a concentration-dependent manner in all cell lines tested after 72 hours of treatment (Figure 3, P ¼ .01). In the PIK3CA mutated cell lines, MFI for p4E-BP1 before treatment ranged from 30e43 although after GDC-0980 treat- ment ranged from 23e25 (P ¼ .03); in the wild-type PIK3CA cell lines p4E-BP1 MFI before treatment ranged from 31e63 vs 15e37 after drug exposure (P ¼ .04) (Figure 3). A similar significant concentration dependent reduction in phosphorylated S6K, another down- stream effector of the mTOR signaling, was also noted for PIK3CA mutated and PI3KCA wild-type cell lines treated with GDC-0980 (data not shown). Represen- tative flow cytometry graph showing reduced p4E-BP1 expression with GDC- 0980 drug treatment can be seen in Figure 4.
COMMENT
In the last few years, the PIK3/AKT/
mTOR pathway has emerged as an important cancer drug target in a variety of human tumors. Class I PI3Ks are het- erodimers comprising a catalytic subunit (p110a, b, d, or g) and a regulatory sub- unit. In normal cells, the PIK3/AKT/
mTOR pathway is tightly controlled, but a large number of cancers harbor genetic changes that result in the deregulated activation of the PIK3/AKT/mTOR pathway that enable tumors to metastasize

of USC patients who may potentially

FIGURE 4
Flow cytometry results of p4-EBP1 after 72 hrs exposure to GDC-0980

Flow cytometry results of p4-EBP1 in representative PI3KCA wild-type USC cell line (ie, uterine serous papillary carcinoma-ARK-2, upper panel ) vs cell line harboring PI3KCA mutation (ie, uterine serous papillary carcinoma-ARK-1, lower panel ) after 72 hrs exposure to GDC-0980. Un- treated cells (light line), cells treated with GDC-0980 1 mM (dotted line), isotype control (continuous line). A significant decrease in p4-EBP1 was consistently detected after the exposure to GDC-0980 in both representative cell lines.
English. GDC-0980, a PIK3CA/mTOR inhibitor, is highly active against USC in vitro. Am J Obstet Gynecol 2013.
benefit from dual PIK3CA/mTORC1/2 inhibitors such as GDC-0980. However, demonstration of the contribution of a mutated/amplified gene to sensitivity to a specific targeted agent requires func- tional validation. Consistent with this view, in this study we have analyzed the presence of oncogenic PIK3CA muta- tions and c-erbB2 gene amplifications in a large series of primary, highly purified USC cell lines. Next, to evaluate whether a mutation profile is predictive of drug response in vitro we have studied the antitumor activity of GDC-0980 in 15 primary USC cell lines established as long-term cultures harboring c-erbB2 gene amplification and wild-type or oncogenic PIK3CA mutations. Consistent with the recent literature derived from comprehensive whole exome sequencing studies,11-13 we found 22.7% of the pri- mary USC cell lines to harbor mutations in exons 9 and 20 of the PIK3CA gene. These 2 exons are known to encompass
>80% of mutations found in human cancers.19 All of the PI3KCA mutations found in our USC cell lines were activating missense mutations. These mutations present in exon 9 and 20 are single amino acid substitutions in the helical or kinase domain, respectively, and result in a gain of enzymatic function, ultimately acti- vating AKT signaling and inducing onco-

and invade normal tissues, continue to grow under conditions of low nutrients or low oxygen, and also resist radiation and chemotherapy treatment.6-10 Targeting the PIK3/AKT/mTOR pathway using potent and selective dual PIK3CA/mTOR inhibitors may therefore represent a compelling therapeutic strategy against biologically aggressive endometrial cancer such as USC.
GDC-0980 has been shown to inhibit both the PI3Ks and mTORC1/mTORC2 complexes and to have remarkable activ- ity in a variety of human cancer cell lines through the inhibition of both PIK3CA and mTOR signaling both in vitro as well as in vivo.14,15 GDC-0980 has also demonstrated efficacy in non-PI3KCA mutated cell lines. It has been postulated and subsequently shown that other downstream molecular effects occur as a consequence of GDC-0980 exposure
in vitro such as a dose-dependent reduc- tion in cyclin D1 levels. As cyclin D1 helps control progression of cells through the cell cycle, this effect of GDC-0980 may explain its effectiveness regardless of mutational status. Also other markers of apoptosis such as cleaved PARP are increased with GDC-0980 administra- tion. On the basis of these promising preclinical results, GDC-0980 is currently being tested in phase II studies in patients with advanced solid tumors including endometrial cancer.
Our group as well as others have recently taken advantage of next genera- tion sequencing technologies to evaluate the mutational landscape of USC.11-13 These studies have revealed amplifica- tion of the HER2/neu gene and onco- genic mutations of the PIK3CA gene in a significant number of USC, high- lighting a rational way to guide selection
genic transformation. The helical and kinase domain mutations are thought to trigger gain of function through different mechanisms. These mutations have been shown to have different requirements for the interaction with PI3K regulatory subunit p85 and with RAS-GTP. Of in- terest, we found the oncogenic PIK3CA mutations to coexist with HER2/neu amplification in 44% of the USC primary cell lines tested. This association between HER2/neu amplification and PIK3CA gene mutations has previously been re- ported in breast cancer cell lines.10 We found GDC-0980 to reduce the growth of all USC cell lines tested. However, the magnitude of the growth reduction detected amongst the different USC cell lines vary significantly according to their HER2/neu gene amplifications and/or PIK3CA mutation status. The PIK3CA/
AKT/mTOR pathway lies downstream

of HER2/neu and PI3KCA/AKT/mTOR pathway aberrant activation secondary to HER2/neu gene amplification has been recently shown to correlate with the sensitivity of a subset of breast cancer cell

pathway in a concentration-dependent manner through reduced phosphoryla- tion of 4-EBP1, a critical downstream target, in all USC cell lines tested. Furthermore, the decrease of 4-EBP1

agent alone in suppressing tumor growth and metastasis. In further support of this targeted therapy with traditional chemo- therapy, a study on endometrial cancer demonstrated that rapamycin in combi-

9,10
lines to mTOR inhibitory drugs.
As
and S6K levels after GDC-0980 treat-
nation with paclitaxel results in increased

such FISHþ cell lines maybe more sensi- tive to GDC-0980 as a result of upstream activation of the PI3KCA/AKT/mTOR pathway with a consequently high cellular reliance on this pathway for cell survival and tumor cell metastasis.
Consistent with these data, as a group, we found USC harboring HER2/neu gene amplifications to be significantly more sensitive to the activityof GDC-0980 when compared with USC cell lines not harboring HER2/neu gene amplification. These data seem to suggest that HER2/neu gene amplified USC may be more depen- dent on continued PI3K/Akt/mTOR pathway activation than HER2/neu- negative tumors. Importantly, PIK3CA mutated cell lines harboring HER2/neu gene amplification were significantly more sensitive to GDC-0980 than USC cell lines
ment was found to be significant in all USC cell lines studied whether USC cell lines harbored the PIK3CA mutation. These results are limited by the relatively small number of primary USC cell lines studied. Investigating a larger number of USC cell lines may be required to determine whether there is a more pronounced decrease in p4-EBP1 levels after GDC-0980 exposure in PI3KCA mutated USC cell lines, which may be at least partially explained by the poten- tially higher activation of the mTOR pathway found in this subset of USC.
Importantly, first generation mTOR inhibitors have been shown to exhibit activity in advanced or recurrent endo- metrial cancer. A phase II study evaluated single-agent activity of temsirolimus, given at 25 mg intravenously weekly in
induction of apoptosis when compared with that of eithercytotoxic agent alone. 24
In conclusion, our report provides the first preclinical demonstration that GDC-0980 may represent a novel, po- tentially effective therapeutic strategy against HER2/neu amplified USC and that the identifi cation of PIK3CA mu- tations in these tumors may be used as a method to potentially select USC pa- tients more likely to respond to GDC- 0980. Although our report is limited by the relative small number of PIK3CA mutated USC cell lines studied, these fi ndings add to the growing preclinical evidence of the effi cacy of PIK3CA/
mTOR inhibition in endometrial cancer cells and suggest that PIK3CA may have a signifi cant role in the treatment of selected and more aggressive endome-

harboring wild-type PIK3CA. To rule out that a fast or slow growth rate of the
patients with recurrent or metastatic endometrial cancer.20 Of 19 patients
trial cancers such as USC.
-

different HER2/neu FISHþ USC cell lines studied was a potential confounding factor in the interpretation of our results we evaluated all PIK3CA mutated and wild- type cell lines tested in kinetics experi- ments. Population doubling times of all the FISHþ USC cell lines studied in these experiments were found to be similar. Taken together, our results suggest for the first time that USC harboring PIK3CA mutations may be physiologically more dependent on this kinase for activation of the mTOR pathway to confer an
evaluated for response, 5 (26%) demon- strated partial response and 12 (63%) showed stable disease as their best response. Response did not correlate with PTEN status as evaluated by immuno- chemistry.20 Another phase II trial of single-agent ridaforolimus (AP23573) including patients with progressive endometrial cancer (23 adenocarcinomas, 5 carcinosarcomas, 6 serous papillary carcinomas, and 1 clear cell carcinoma) showed a 33% response rate with a low incidence of adverse effects.21 In a recent
ACKNOWLEDGMENT
We wish to thank Genentech-Roche, South San Francisco, CA for providing GDC-0980 free of charge for our experiments.

REFERENCES
1.Siegel R, Naishadham P, Jemal A. Cancer statistics, 2013. CA Cancer J Clin 2013;63:11-30.
2.Bokhman JV. Two pathogenetic types of endometrial carcinoma. Gynecol Oncol 1983; 15:10.
3.Hendrickson M, Ross J, Eifel P, et al. Uterine papillary serous carcinoma: a highly malignant form of endometrial adenocarcinoma. Am J

increased growth and survival advantage.
22
study Bae-Jump et al,
it was demon-
Surg Pathol 1982;6:93-108.

Our results obtained using GDC-0980 against uterine serous tumors are there- fore in agreement with the data reported from research in Her2/neu amplified breast cancer cell lines, which also observed a greater dependency of mutant cell lines to the canonical PIK3CA/AKT/
mTOR pathway.9,10
Next, we evaluated the ability of GDC- 0980 to induce a downstream cellular response by measuring phosphorylation of the 4-EBP1 protein by flow- cytometry. In this regard, we showed that GDC-0980 inhibited the mTOR
strated that rapamycin has activity in fresh endometrial tumor explants independent of PTEN and Akt status suggesting that mTOR inhibitors may be a useful targeted therapy for both type I and type II endometrial cancers. Finally, the potential use of mTOR inhibitors in combination with traditional chemotherapeutic agents is also a promising treatment option. Zeng et al23 evaluated response to rapamycin and cyclophosphamide in a murine model of triple-negative breast cancer. They showed that this combination ach- ieved improved efficacy relative to either
4.Hamilton CA, Cheun MK, Osann K, et al. Uterine papillary serous and clear cell carci- nomas predict for poorer survival compared to grade 3 endometrioid corpus cancers. Br J Cancer 2006;94:642-6.
5.Schwartz PE. The management of serous papillary uterine cancer. Curr Opin Oncology 2006;18:494-9.
6.Dancey JE. Therapeutic targets: MTOR and related pathways. Cancer Biol Ther 2006;5: 1065-73.
7.Vanhaesebroeck B, Guillermet-Guibert J, Graupera M, Bilanges B. The emerging mech- anisms of isoform-specific PI3K signalling. Nat Rev Mol Cell Biol 2010;11:329-41.
8.English DP. Roque DM. Santin AD. HER2 expression beyond breast cancer: therapeutic

implications for gynecologic malignancies. Mol Diagn Ther 2013;17:85-99.
9.She QB, Chandarlapaty S, Ye Q, et al. Breast tumor cells with PI3K mutation or HER2 ampli- fi cation are selectively addicted to Akt signaling. PLoS ONE 2008;3:e3065.
10.Brachmann SM, Hofmann I, Schnell C, et al. Specific apoptosis induction by the dual PI3K/
mTor inhibitor NVP-BEZ235 in HER2 amplified and PIK3CA mutant breast cancer cells. PNAS USA 2009;106:22299-304.
11.Kuhn E, Wu RC, Guan B. Identification of molecular pathway aberrations in uterine serous carcinoma by genome-wide analyses. J Nat Cancer Inst 2012;104:1503-13.
12.Le Gallo M, O’Hara AJ, Rudd ML, et al. Exome sequencing of serous endometrial tu- mors identifies recurrent somatic mutations in chromatin-remodeling and ubiquitin ligase complex genes. Nat Genet 2012;44:1310-5.
13.Zhao S, Choi M, Overton JD, et al. Land- scape of somatic single-nucleotide and copy- number mutations in uterine serous carcinoma. PNAS USA 2013;110:2916-21.
14.Salphati L, Pang J, Plise EG, et al. Preclinical assessment of the absorption and disposition of

the phosphatidylinositol 3-kinase/mammalian target of rapamycin inhibitor GDC-0980 and prediction of its pharmacokinetics and efficacy in human. Drug Metab Dispos 2012;40:1785-96.
15.Wallin JJ, Edgar KA, Guan J, et al. GDC- 0980 is a novel class I PI3K/mTOR kinase in- hibitor with robust activity in cancer models driven by the PI3K pathway. Mol Cancer Ther 2011;10:2426-36.
16.Cross SN, Cocco E, Bellone S, et al. Differ- ential sensitivity to platinum-based chemotherapy in primary uterine serous papillary carcinoma cell lines with high vs low HER-2/neu expression in vitro. Am J Obstet Gynecol 2010;203:162. e1-8.
17.Roque DM, Bellone S, English DP, et al. Tubulin-b-III overexpression by uterine serous carcinomas is a marker for poor overall survival after platinum/taxane chemotherapy and sensi- tivity to epothilones. Cancer 2013;119:2582-92.
18.Kotecha N, Flores NJ, Irish JM, et al. Single- cell profiling identifies aberrant STAT5 activation in myeloid malignancies with specific clinical and biologic correlates. Cancer Cell 2008;14:335-43.
19.Forbes SA, Bhamra G, Bamford S, et al. The catalogue of somatic mutations in cancer

(COSMIC). Curr Protoc Hum Genet Suppl 2008;57. 10.11.1-10.11.26.
20.Oza AM, Elit L, Biagi J, et al. Molecular correlates associated with a phase II study of temsirolimus (CCI-779) in patients with meta- static or recurrent endometrial cancer – NCI- CIND 160. J Clin Oncol 2006;24:121.
21.Colombo N, McMeekin S, Schwartz P, et al. A phase II trial of the mTOR inhibitor AP23573 as a single agent in advanced endometrial cancer. J Clin Oncol 2007;25:5516.
22.Bae-Jump VL, Zhou C, Boggess JF, et al. Rapamycin inhibits cell proliferation in type I and type II endometrial carcinomas: a search for biomarkers of sensitivity to treatment. Gynecol Oncol 2010;119:579-85.
23.Zeng Q, Yang Z, Gao YJ, et al. Treating triple-negative breast cancer by a combination of rapamycin and cyclophosphamide: an in vivo bioluminescence imaging study. Eur J Cancer 2010;46:1132-43.
24.Shafer A, Zhou C, Gehrig PA, et al. Rapa- mycin potentiates the effects of paclitaxel in endometrial cancer cells through inhibition of cell proliferation and induction of apoptosis. Int J Cancer 2010;126:1144-54.