Therapeutic Silencing of KRAS Using Systemically Delivered siRNAs

Since its discovery over 30 years ago (1, 2), the KRAS proto-oncogene has remained the single most elusive cancer target. Despite the vast heterogeneity of all malignancies, mutational activation of the RAS GTPases (HRAS, NRAS, and KRAS) is present in approximately one third of all cancers (3). Although these small monomeric GTPases are part of a superfamily of more than 150 members, direct mutation of other members is rare (4). Typically, RAS activation is catalyzed by guanine nucleotide exchange factors (GEF) to a GTP-bound state, and subsequently hydrolyzed by GTPase-activating proteins (GAP) to its inactive, GDP-bound state. However, KRAS missense mutations in codons 12, 13, or 61 sterically interfere with GAP hydrolysis, leading to constitutive activation and promotion of many cancer hallmarks, such as cellular proliferation, survival, cytoskeletal reorganization, and motility (5). Although valiant attempts have been made to develop pharmaceutical inhibitors of mutant KRAS-driven cancers, KRAS itself is still widely regarded as “undruggable.”

Since the first report of RNAi in 1998 (6), there has been an explosion in efforts to use such a strategy for therapeutic gain (7). Therapeutic RNAi is especially attractive because it enables silencing of cancer molecular targets that otherwise may not be inhibited using conventional approaches. Although competitive ATP kinase inhibitors (e.g., imatinib) or mAbs (e.g., trastuzumab) have revolutionized treatment of some cancers (8, 9), the lack of such success in KRAS targeting prompted us to investigate whether RNAi has therapeutic potential for drug development. Here, we report a proof-of-concept study in lung and colon cancer preclinical models that demonstrates the efficacy of KRAS silencing using nanoparticle-mediated siRNA delivery. In addition, we demonstrate in several models that KRAS silencing in vivo can potently inhibit development of metastatic disease, the cause of death in approximately 90% of patients with cancer (10).

All cell lines were maintained in 5% CO₂/95% air at 37°C. Lung (A549 and H1299) and ovarian (RMUG-S) cells were obtained from the ATCC and maintained in RPMI 1640 supplemented with 10% FBS and 0.1% gentamicin sulfate (GeminiBioproducts). The A549-Luciferase cell line was made following stable transduction with lenti-virus carrying the luciferase gene (the lentiviral vector was kindly provided by Craig Logsdon's lab). The HCP1 colon cell lines were obtained from a human-derived xenograft model at the M.D. Anderson Cancer Center under an Institutional Review Board–approved protocol as recently described (11). Cell lines were routinely tested to confirm the absence of Mycoplasma, and all in vitro experiments were conducted with 60% to 80% confluent cultures. All cells were reverse-transfected with RNAiMax reagent (Invitrogen) using siRNA molecules (Sigma) at a final concentration of 10 to 20 nmol/L. Media were changed 4 hours following transfections to minimize toxicity. The siRNA sequences used for KRAS siRNA experiments are as follows:

• Negative control: sense UUCUCCGAACGUGUCACGU

• Antisense ACGUGACACGUUCGGAGAA

• Seq #1: sense GUGCAAUGAAGGGACCAGUA,

• Antisense UACUGGUCCCUCAUUGCAC

• Seq #2: sense GUCUCUUGGAUAUUCUCGA,

• Antisense UCGAGAAUAUCCAAGAGAC

• Seq #3: Sense CAGCUAAUUCAGAAUCAUU,

• Antisense AAUGAUUCUGAAUUAGCUG.

Female athymic nude mice were purchased from the NCI, Frederick Cancer Research and Development Center (Frederick, MD). These animals were cared for according to guidelines set forth by the American Association for Accreditation of Laboratory Animal Care and the U.S. Public Health Service policy on Human Care and Use of Laboratory Animals. All mouse studies were approved and supervised by the M.D. Anderson Cancer Center Institutional Animal Care and Use Committee. All animals used were between 8 to 12 weeks of age at the time of injection. For all animal experiments, cells were trypsinized, washed, and resuspended in Hanks' balanced salt solution (HBSS; Gibco) before injection. For the orthotopic lung cancer model, A549-Luc cells were injected by an intrapulmonary technique [7.5 × 10⁵ in 100 μL 1:1 mixture of HBSS and BD Matrigel (BD Biosciences)] as previously described (12). For the intrapulmonary injections, mice were anesthetized with ketamine + xylazine and placed in the right lateral decubitus position. Following skin cleaning with an alcohol swab, an incision parallel to the rib cage between ribs 10 and 11 was made to visualize the lung through the thorax. A 1-mL tuberculin syringe with a 30-g needle was used to inject the cell suspension directly into the lung parenchyma at the left lateral dorsal axillary line. After injection, the incision was closed using surgery clips, and the mice were turned on the left lateral decubitus until fully recovered.

For both colon cancer models, 2.5 × 10⁵ HCP1 cells per mouse in 100 μL HBSS were either injected subcutaneously or intrasplenic (experimental liver metastases model). In the liver metastases model, mice were anesthetized under isofluorane for splenic isolation and cell line injection (day 1), as well as the following day after injection (day 2) to perform splenectomy (11). Treatment continued until mice in any group became moribund (approximately 3 weeks).

For the A549 xenograft time-kinetic experiment used to assess for KRAS knockdown following siRNA delivery, 2.5 × 10⁶ A549 parental cells were injected subcutaneously into athymic nude mice. Mice were randomly assigned to receive either negative control (NC) siRNA or the combination of KRAS siRNAs Seq #2 and #3 (150 μg siRNA/kg/mouse). After 4 weeks of cell line injection, siRNAs packed in 1,2-dioleoyl-sn-glycero-3-phosphatidylcholine (DOPC) nanoliposomes were delivered intraperitoneally, and one tumor per group was obtained at 24, 48, and 96 hours following a single delivery.

For all therapeutic experiments, mice were randomly divided and assigned to their respective treatment groups (n = 10 mice/group). A dose of 150 μg siRNA/kg/mouse was packaged within DOPC nanoliposomes and delivered intraperitoneally at twice weekly intervals as previously described (13). For the A549 lung cancer model, treatment began 10 days after cell line injection and continued for approximately 4 weeks. A 160 μg dose of pharmaceutical grade cisplatin (CDDP; 1 mg/mL concentration) was administered to respective groups once weekly intraperitoneally. Weekly imaging was performed using the Xenogen IVIS 200 system within 10 minutes following injection of D-Luciferin (150 μg/mL). Living image 2.5 software was used to determine the regions of interest (ROI), and average photon radiance (p/s/cm²/sr) was measured for each mouse.

For the subcutaneous colon cancer model, tumors were monitored until they measured approximately 100 mm³, as determined using the following formula: volume = (L × W × W)/2, where L equals the greatest dimension of the tumor and W equals the perpendicular measurement of the tumor. Twice-weekly treatment began once tumors measured approximately 100 mm³ and continued until tumors in any group were approximately 15 mm in any dimension. In the time-kinetic experiment, 2 weeks following HCP1 cell line injection into a total of 4 mice, mice were randomly assigned to receive either NC siRNA-DOPC or KRAS siRNA (Seq #3)-DOPC. After a single injection, mice in each group were sacrificed, and tumors were snap frozen at 48 and 96 hours following the injection.

In all experiments, once mice in any group became moribund, they were all sacrificed, necropsied, and tumors were harvested. Tumor weights, number, and location of tumor nodules were recorded. In the A549 orthotopic lung cancer model, tumor volume was determined using calipers as follows: volume = (L × W × W)/2, where L equals the greatest dimension of the tumor and W equals the perpendicular measurement of the tumor. Tumor tissue was either fixed in formalin for paraffin embedding, frozen in optimal cutting temperature (OCT) media to prepare frozen slides, or snap frozen for lysate preparation.

siRNA for in vivo intratumor delivery was incorporated into DOPC. DOPC and siRNA were mixed in the presence of excess tertiary butanol at a ratio of 1:10 (w/w) siRNA/DOPC (Supplementary Fig. S1). Tween 20 was added to the mixture in a ratio of 1:19 Tween 20:siRNA/DOPC. The mixture was vortexed, frozen in an acetone/dry ice bath, and lyophilized. Before in vivo administration, this preparation was hydrated with PBS at room temperature at a concentration of 150 μg siRNA/kg per injection (each mouse received 200 μL of DOPC:siRNA:PBS solution by the intraperitoneal route).

Lysates from cultured cells were prepared using modified RIPA buffer [50 mmol/L Tris–HCl (pH 7.4), 150 mmol/L NaCl, 1% Triton, 0.5% deoxycholate] plus 25 μg/mL leupeptin, 10 μg/mL aprotinin, 2 mmol/L EDTA, and 1 mmol/L sodium orthovana. To prepare lysates of snap-frozen tissue from mice, approximately 30 mm³ cuts of tissue were disrupted with a tissue homogenizer and centrifuged at 13,000 rpm for 30 minutes within modified RIPA buffer. The protein concentrations were determined using a BCA Protein Assay Reagent kit (Pierce Biotechnology). Lysates (100 μg/lane) were loaded and separated on 8% sodium dodecyl sulfate–polyacrylamide gels. Proteins were transferred to a nitrocellulose membrane by semidry electrophoresis (Bio-Rad Laboratories) overnight, blocked with 5% milk for 1 hour, and then incubated at 4°C with primary antibody [KRAS, 1:500, Abcam ab81075 (cell lysates); KRAS, 1:500, Santa Cruz SC-30 (tumor lysates); pERK1/2, 1:1,000, Cell Signaling (4695); ERK1/2, 1:1,000, Cell Signaling (4376); pMEK1/2, 1:1,000, Cell Signaling (2338); MEK1/2, 1:1,000, Cell Signaling (4694); Vinculin, 1:1,000, Santa Cruz SC-25336] overnight. After washing with TBST, the membranes were incubated with horseradish peroxidase (HRP)–conjugated horse anti-mouse IgG (1:2,000; GE Healthcare) for 2 hours. HRP was visualized by use of an enhanced chemiluminescence detection kit (Pierce). To confirm equal sample loading, the blots were probed with an antibody specific for Vinculin. Densitometry was calculated using ImageJ software.

For mRNA quantification, total RNA was isolated by using a Qiagen RNeasy kit. Using 500 ng of RNA, cDNA was synthesized by using a Verso cDNA kit (Thermo Scientific) as per the manufacturer's instructions. Analysis of mRNA levels was performed on a 7500 Fast Real-Time PCR System (Applied Biosystems) with SYBR Green–based real-time PCR. Specific primers for KRAS F-TGACCTGCTGTGTCGAGAAT, R-TTGTGGACGAATATGATCCAA; 18S F-CGCCGCTAGAGGTGAAATTC, R-TTGGCAAATGCTTTCGCTC were used; 18S rRNA was used as a housekeeping gene. PCR was done with reverse-transcribed RNA and 100 ng/μL of sense and antisense primers in a total volume of 20 μL. Each cycle consisted of 15 seconds of denaturation at 95°C and 1 minute of annealing and extension at 60°C (40 cycles).

Cell viability assays were performed by testing cell's ability to reduce the tetrazolium salt [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt] to a formazan. Briefly, cells were seeded in a 96-well plate and were reverse transfected with either NC siRNA, KRAS siRNA (Seq #1, Seq #2, Seq #3 or the combination of Seq #2+#3). At 24, 48, 72, 96, and 120 hours following transfection, cells were incubated with 0.15% 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) for 2 hours at 37°C. In the A549 experiment, 10 to 20 nmol/L of siRNA was used as indicated, and in the H1299 and RMUG-S cell lines, 20 nmol/L siRNA was used. The supernatant was removed, cells were dissolved in 100-μL DMSO, and the absorbance at 570 nm was recorded.

Staining was performed in formalin-fixed, paraffin-embedded tumor sections (8 μm thickness) or from OCT-embedded frozen tissue sections. After deparaffinization, rehydration, and antigen retrieval or fixation, 3% H₂O₂ was used to block the endogenous peroxidase activity for 10 minutes. Protein blocking of nonspecific epitopes was done using either 5% normal horse serum, 1% normal goat serum, or 2.8% fish gelatin in PBS for 20 minutes. Slides were incubated with primary antibody for Ki-67 (rabbit anti-mouse, 1:200, Abcam ab15580) or cleaved caspase-3 (BioCare Medical, 1:100) overnight at 4°C. Terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick-end labeling (TUNEL) staining was performed using the Promega Kit (Promega), and CD31 staining to assess microvessel density was performed as previously described (12, 14). For immunohistochemistry, after washing with PBS, the appropriate amount of HRP-conjugated secondary antibody was added and visualized with 3,3′-diaminobenzidine chromogen and counterstained with Gill's hematoloxylin #3. Light field images were obtained using a Nikon Microphot-FXA microscope and Leica DFC320 digital camera. Proliferation and apoptotic indices were determined using 6 to 10 representative fields at ×200 magnification for each tumor, whereas microvessel density was determined at ×100 magnification (5 tumors per group). Ki-67 and cleaved caspase-3–positive cells per high-powered field were enumerated using Cell Profiler 2.0 software, whereas TUNEL-positive cells per high-powered field were enumerated manually (15). Quantification was performed in a blinded fashion.

For animal experiments, 10 mice were assigned per treatment group. This sample size provided 80% power to detect a 50% reduction in tumor mass with 95% confidence. Mouse and tumor weights and the number of tumor nodules for each group were compared using the Student t test (for comparisons of two groups) and the ANOVA (for multiple group comparisons). For values that were not normally distributed (as determined by the Kolmogrov–Smirnov test), the Mann–Whitney rank sum test was used. A P value ≤0.05 was deemed statistically significant. All statistical tests were two-sided and were performed using GraphPad Prism 6.0.

To test the efficacy of KRAS silencing in vivo, we first searched for siRNA sequences that potently inhibit KRAS expression regardless of the specific missense mutation at codons 12, 13, or 61. Low concentrations (20 nmol/L) of two siRNA sequences (Seq #2 and Seq #3) within the coding sequence were found to inhibit KRAS mRNA expression in mutant (A549 lung adenocarcinoma, KRAS^G12S) and wild-type (RMUG-S ovarian carcinoma) cell lines (Fig. 1A and B). Western blot analyses further confirmed greater than 90% knockdown of KRAS protein in mutant (A549) and wild-type (H1299, RMUG-S) cancer cell lines (Fig. 1C). Compared with NC siRNA, a single transfection of Seq #2 or Seq #3 siRNAs significantly inhibited cell viability in the A549 cell line, but not in the wild-type cells (Fig. 1D), suggesting attenuated growth is not due to off-target effects of the siRNA sequences and is consistent with A549 having a KRAS oncogene “addiction” (16). Furthermore, the combination of Seq #2 and Seq #3 at low concentrations (10 nmol/L or 20 nmol/L) led to substantial reductions in cell viability (Fig. 1D). As compared with H1299, KRAS siRNA sequences #2 and #3 in A549 led to marked reductions in pERK and pMEK, molecules involved in the downstream signaling of the MAPK pathway (Fig. 1E).

Using our well-characterized DOPC nanoliposomes (Supplementary Fig. S1; ref. 13), we have recently demonstrated highly effective systemic delivery of oligonucleotides in orthotopic lung cancer models (12). Through exploitation of the enhanced-permeability and retention (EPR) effect (7), we also demonstrated the ability to systemically deliver siRNAs to metastatic lesions (12). Thus, for all in vivo experiments, we assessed therapeutic efficacy of KRAS targeting with DOPC-mediated siRNA delivery.

We first assessed whether systemically delivered KRAS siRNAs can knock down KRAS expression in vivo. Four weeks following subcutaneous injection of 2.5 × 10⁶ A549 cells, compared with NC siRNA, following a single injection of KRAS siRNAs Seq #2 and #3, A549 tumors had sustained knockdown of KRAS protein expression for up to 96 hours (Fig. 2A). We have previously observed similar patterns of knockdown with other targets, with rebound typically occurring around 5 to 6 days after delivery (13, 17). Next, to evaluate longitudinal effects of KRAS siRNA treatment, we used luciferase-labeled A549 (A549-Luc) cells to enable IVIS imaging. Furthermore, because cisplatin (CDDP) is the backbone of lung cancer treatment, we assessed whether chemotherapy and KRAS targeting would have enhanced antitumor effects. In addition to primary tumors, this model demonstrated the ability to form mediastinal lymph node and chest wall metastases (Fig. 2B).

One week following orthotopic injection of 7.5 × 10⁵ A549-Luc cells, mice were randomly assigned to the following treatment groups (n = 10 mice/group): (i) NC siRNA-DOPC, (ii) KRAS siRNA-DOPC, (iii) NC siRNA-DOPC+CDDP, and (iv) KRAS siRNA-DOPC+CDDP. For the KRAS siRNA treatment groups, Seq #2 and Seq #3 were combined. For the CDDP treatment groups, 160 μg of CDDP was given intraperitoneally once weekly. One week after cell line injection, twice-weekly treatment commenced and weekly imaging was performed (Fig. 2C). Following four weeks of systemic therapy, compared with control siRNA treatment, mice in the KRAS siRNA treatment group had an 81% reduction in luminescence (P < 0.01), whereas the addition of CDDP did not lead to significant reductions in the NC siRNA+CDDP or KRAS siRNA+CDDP groups (Fig. 2D and E). Longitudinal assessments of luciferase activity demonstrated significant growth inhibition from KRAS siRNA treatment (Fig. 2F), and rapid disease progression during the last week of treatment was significantly attenuated as compared with NC siRNA (Fig. 2G). Following four weeks of treatment, all mice were sacrificed, and an extensive necropsy of gross tumor burden was quantified. As compared with control siRNA, significant reductions in primary tumor size (KRAS siRNA: 50%, P = 0.003; NC siRNA+CDDP: 6%, P = 0.24; KRAS siRNA+CDDP: 73%, P = 0.008) and aggregate mass of distant metastases (KRAS siRNA: 77%, P = 0.0002; NC siRNA+CDDP: 11%, P = 0.85; KRAS siRNA+CDDP: 74%, P = 0.009) were only observed in the KRAS siRNA treatment groups (Fig. 2H and I and Supplementary Fig. S2). The number of distant metastases was only significantly decreased in the KRAS siRNA+CDDP group (Fig. 2J). Intriguingly, only the KRAS siRNA treatment groups had significantly less frequent mediastinal metastases, whereas NC siRNA+CDDP did not (Fig. 2K). Taken together, systemically delivered KRAS siRNAs demonstrated significant therapeutic activity on the primary tumor as well as control of metastatic spread, while the addition of cytotoxic chemotherapy was not additive.

To further assess the biologic effects of systemic KRAS siRNA treatment, we stained the tumor samples for proliferation, apoptosis, and angiogenesis markers. Compared with control siRNA, significant increases in caspase-3 activity and TUNEL staining were observed in all treatment groups, with the greatest effects seen in the KRAS siRNA+CDDP group (Fig. 3A and B). Compared with control siRNA, significant reductions in Ki-67 indices were seen in all treatment groups (KRAS siRNA: 34%; NC siRNA+CDDP: 52%; KRAS siRNA+CDDP: 36%; all P < 0.0001; Fig. 3C). Also, compared with control siRNA, all treatment groups had significant reductions in microvessel density (KRAS siRNA: 61%; NC siRNA+CDDP: 55%; KRAS siRNA+CDDP: 76%; all P < 0.001; Fig. 3D). Compared with NC siRNA+CDDP, KRAS siRNA+CDDP had a 45% reduction in microvessel density (P < 0.05, Fig. 3D). These data indicate that in vivo silencing of KRAS in tumors leads to induction of apoptosis and inhibition of proliferation and angiogenesis. Although the combination of KRAS siRNA with cisplatin had additive biologic effects on the tumors (Fig. 3A–D), there were no appreciable effects on overall disease burden at the time of necropsy. Although these findings suggest combining KRAS silencing and cisplatin may not have additive therapeutic effects, the dose chosen may have been subtherapeutic.

Next, using a recently characterized patient-derived colon cancer cell line with a KRAS^G12D mutation (11), HCP1, we assessed whether downstream mediators of KRAS signaling are abrogated following KRAS siRNA. treatment. Three weeks after early passage HCP1 subcutaneous tumors were created, mice were randomly assigned to receive a single injection of either NC siRNA-DOPC or KRAS siRNA-DOPC. Similar to before (Fig. 2C), at both 48- and 96-hour timepoints, clear reductions in KRAS and pERK signaling (Fig. 4A) were observed.

Next, we assessed whether systemic KRAS siRNA could inhibit tumor growth in a colon cancer model. Following subcutaneous injection of 2.5 × 10⁵ HCP1 cells, tumors were monitored until reaching approximately 100 mm³. Mice were randomly assigned to the following treatment groups (n = 10 mice/group): (i) NC siRNA-DOPC or (ii) KRAS siRNA-DOPC. Twice-weekly treatments were given and tumor volumetric assessments were made with calipers every three days. As soon as one week after treatment, significant reduction in tumor growth was noted in the KRAS siRNA-DOPC group, and following two weeks of treatment, tumors were 68.6% smaller (P = 0.033) than the NC siRNA group (Fig. 4B).

Because the predominant site of distant metastasis in patients with colon cancer is the liver, we assessed whether KRAS siRNA would be effective in an experimental metastasis model. Splenectomy was performed one day following intrasplenic injection of 2.5 × 10⁵ HCP1 cells, and mice were randomly assigned to the following groups (n = 10 mice/group): (i) NC siRNA-DOPC or (ii) KRAS siRNA-DOPC. As before, for the KRAS siRNA treatment groups, Seq #2 and Seq #3 were combined. Systemic treatment began 5 days following cell line injection and continued until mice in either group became moribund (22 days). Following nearly 3 weeks of treatment, KRAS siRNA led to a 73% reduction in aggregate metastatic disease (P = 0.014; Fig. 4C) and a trend toward a reduction in the number of gross metastatic liver lesions (P = 0.09; Fig. 4D and E). Notably, no mice in any of the therapeutic experiments exhibited any overt signs of toxicity from KRAS siRNA treatment, such as decreased feeding habits, activity, or total body weight. Taken together, systemically delivered KRAS siRNAs significantly inhibited primary tumor growth and outgrowth of distant metastases in lung and colon cancer models driven by mutant KRAS.

Multiple approaches have been attempted to target mutant RAS signaling; however, to date, no pharmacologic inhibitors are currently in clinical use. Although the ability to develop ATP-competitive inhibitors with nanomolar affinities has led to clinical efficacy (8), the low picomolar binding affinity of small GTPases for GTP and milimolar cellular concentrations of GTP make this approach far less feasible (4). Furthermore, the conformational changes following substitution of glycine (only amino acid without a side chain) for other residues in the “finger loop” sterically interferes with GAP activity, making development of GAP-mimics unlikely (18). Although targeting the prenylation process of KRAS with farnesyl transferase inhibitors showed initial preclinical promise (19), enthusiasm was dampened by lack of efficacy in clinical trials (20). Most recent efforts have turned toward synthetic-lethality approaches as well as targeting of downstream signaling mediators of KRAS pathways (21, 22). Although such approaches have shown preclinical efficacy (23), compensatory mechanisms, toxicity from multiple inhibitors, and ultimately efficacy in clinical trials are important hurdles that remain.

Using mutant KRAS models of lung and colon cancer, we demonstrate the therapeutic potential of systemically delivered RNAi as a novel treatment modality. Although therapeutic effects were seen through decreased downstream signaling and inhibition of proliferation, we also observed significant decreases in the frequency and burden of distant metastases. KRAS siRNA delivery as a therapeutic approach has many characteristics that make it desirable for further drug development. First, so far the first-in-human clinical trials using nanoparticle-mediated siRNA delivery have been safe, well tolerated, and in some cases have led to complete pathologic responses in metastatic disease (24, 25). Second, although in this proof-of-concept study, siRNA sequences that targeted all forms of mutant KRAS (and wild-type) were used, siRNA sequences that preferentially target only the mutant allele could be developed, thus theoretically avoiding toxicity from targeting the wild-type protein in normal tissues. Lastly, although other current approaches rely upon blocking the effects of mutant KRAS, whose downstream mediators consist of a highly complex circuitry, use of RNAi aims at eliminating the functional capstone, KRAS itself. Although therapeutic RNAi has important challenges of its own, such as developing targeted platforms that are biocompatible and potently inhibit target expression, the field is rapidly moving forward and turning this approach into a reality (7).

L.M. Ellis is a consultant/advisory board member for Genentech/Roche, ImClone/Lilly and Amgen. No potential conflicts of interest were disclosed by the other authors.

Conception and design: C.V. Pecot, S.Y. Wu, H.J. Dalton, G. Lopez-Berestein, L.M. Ellis, A.K. Sood

Development of methodology: C.V. Pecot, S.Y. Wu, C. Rodriguez-Aguayo, V. Gonzalez-Villasana, A.K. Sood

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): C.V. Pecot, S.Y. Wu, J. Filant, R. Rupaimoole, T. Hisamatsu, R. Bhattacharya, A. Maharaj, S. Azam, C. Rodriguez-Aguayo, T.A. Waugh, B. Zand, H.J. Dalton, S. Kopetz, G. Lopez-Berestein, A.K. Sood

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): C.V. Pecot, S.Y. Wu, J. Filant, A.S. Nagaraja, K.M. Gharpure, V. Gonzalez-Villasana, S. Kopetz, G. Lopez-Berestein, A.K. Sood

Writing, review, and/or revision of the manuscript: C.V. Pecot, S.Y. Wu, R. Bhattacharya, A.S. Nagaraja, M.P. Morelli, K.M. Gharpure, H.J. Dalton, S. Kopetz, G. Lopez-Berestein, L.M. Ellis, A.K. Sood

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): C.V. Pecot, S. Bellister, T.A. Waugh, A.K. Sood

Study supervision: C.V. Pecot, M.P. Morelli, G. Lopez-Berestein, L.M. Ellis, A.K. Sood

The authors thank Donna Reynolds for expertise in immunohistochemistry.

Portions of this work were supported by the NIH (CA016672, CA109298, P50 CA083639, P50 CA098258, CA128797, U54 CA151668, CA177909, and U24CA143835), the Cancer Prevention Research Institute of Texas (RP110595), the Ovarian Cancer Research Fund, Inc. (Program Project Development Grant), the Department of Defense (OC073399, W81XWH-10-1-0158), the RGK Foundation, the Gilder Foundation, the Blanton-Davis Ovarian Cancer Research Program, and the Betty Anne Asche Murray Distinguished Professorship (to A.K. Sood). C.V. Pecot was supported by a grant from the NCI (T32 training grant CA009666), the 2011 Conquer Cancer Foundation ASCO Young Investigator Award, the Ben F. Love Fellowship in Innovative Cancer Therapies, the Jeffrey Lee Cousins Fellowship in Lung Cancer Research, and the DoCM Advanced Scholar Program. S.Y. Wu was supported by Ovarian Cancer Research Funds, Foundation for Women's Cancer, and Cancer Prevention Research Institute of Texas training grants (RP101502 and RP101489). S. Kopetz was in part supported by NIH grant CA172670. S. Azam was supported in part by a grant from the National Institute of General medical Sciences under award 5T32 GM007092. L.M. Ellis was in part supported by NIH grant R01CA157880, DOD CA100879, and the William C. Liedtke, Jr., Chair in Cancer Research.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.