Characterization of the Activity of the PI3K/mTOR Inhibitor XL765 (SAR245409) in Tumor Models with Diverse Genetic Alterations Affecting the PI3K Pathway Free

Class I PI3 kinases convert phosphatidylinositol 4,5-bisphosphate (PIP2) to phosphatidylinositol 3,4,5-trisphosphate (PIP₃) in response to external cell stimuli (1, 2). Activation of class I_A PI3Ks (phosphoinositide 3-kinases; PI3Kα, -β, and -δ) is mediated by receptor tyrosine kinases (RTK). G-protein–coupled hormone receptors are implicated in activation of PI3Kβ and class I_B PI3K (PI3Kγ; ref. 3). Ras, another important mediator of extracellular stimuli, can also promote PI3K activation, and PI3K can mediate cellular transformation by Ras (4). Downstream effectors of PI3K signaling, such as phosphoinositide-dependent kinase-1 (PDK1) and AKT, bind to PIP₃ at the cell membrane and are subsequently activated by phosphorylation (1). In turn, PDK1 and AKT activate growth pathways, inhibit apoptotic signaling, and regulate transition through restriction points in the cell cycle via phosphorylation of their respective substrates (1).

mTOR is the kinase component of two multisubunit complexes called mTORC1 (includes mTOR/Raptor) and mTORC2 (includes mTOR/Rictor; refs. 5, 6). mTORC1 is activated via PI3K pathway signaling and also via PI3K-independent mechanisms involving sensing of cellular amino acid levels, AMP levels, and hypoxia, and drives cellular growth by regulating protein translation and degradation (5, 6). mTORC2 is activated via growth factor–dependent signaling via mechanism(s) that are still being elucidated, and regulates cell growth, proliferation, and survival via phosphorylation of the AKT kinase (5–7).

Dysregulation of PI3K pathway components, resulting in hyperactivated PI3K and/or mTORC1 signaling, is observed in various cancers and correlates with tumor growth and survival (1). For example, the catalytic subunit of PI3Kα (p110α), encoded by the PIK3CA gene, is mutated in 12% of human cancers (8). This is likely an underestimate because many of these data are presumably generated by hotspot sequencing. In addition, the tumor suppressor PTEN, which serves as a critical negative regulator of PI3K signaling by converting PIP₃ back to PIP2, is frequently deleted or downregulated in human tumors (1, 9). Moreover, the tumor suppressor gene LKB-1, which negatively regulates mTORC1, is mutated/inactivated in a variety of familial and sporadic tumors (10). PI3K pathway signaling is implicated in tumor cell invasion, migration, and dissemination (11). Genetic and pharmacologic approaches have demonstrated that PI3K signaling mediates VEGF production, neoangiogenesis, vascular permeability, and vessel integrity in preclinical tumor models (12–14).

Resistance to a variety of anticancer therapies, including RTK inhibitors and genotoxic agents, has been attributed to ongoing activation of the PI3K/PTEN pathway (15–17). PI3K pathway inhibitors have been shown to sensitize cancer cells to agents targeting HER2, MET, and EGFR as well as platinum drugs and taxanes (18–23).

mTOR has been extensively explored as an oncology target (24). The macrolide antibiotic rapamycin is a potent inhibitor of the mTORC1 complex. Several analogs of rapamycin have been tested clinically in various oncology indications, and evidence of therapeutic benefit has been observed (24). For example, the U.S. Food and Drug Administration has approved TORISEL (temsirolimus; Pfizer) for patients with advanced renal cell carcinoma (RCC) and Afinitor (everolimus; Novartis) for the treatment of patients with advanced RCC after failure of treatment with sunitinib or sorafenib, and for advanced estrogen receptor–positive breast cancer after failure of treatment with a nonsteroidal aromatase inhibitor. However, the efficacy of these agents may be limited by the fact that rapamycin analogs do not inhibit mTORC2 (7). In addition, mTORC1 inhibition may enhance cell survival by upregulating PI3K/AKT signaling via inhibition of a mTORC1-dependent negative feedback loop acting through PI3K (24). Thus, selective inhibitors of PI3K and mTOR signaling have therapeutic potential as single agents and in combination with other therapies for a variety of cancer indications and several such agents have entered clinical testing in recent years (1, 24).

XL765 (SAR245409) is a potent and selective inhibitor of class I PI3Ks. In addition, XL765 also inhibits mTOR. In cellular assays, treatment with XL765 inhibits phosphorylation of proteins downstream of PI3K and mTOR, including AKT and ribosomal protein S6 (S6RP), in multiple tumor cell lines with diverse molecular alterations affecting the PI3K pathway. In a broad panel of tumor cell lines XL765 inhibits proliferation with a wide range of potencies, which seemed to be influenced by genetic background. Oral administration of XL765 in human xenograft tumor models in athymic nude mice results in dose-dependent inhibition of PI3K pathway components with a duration of action of at least 24 hours. As a single agent, XL765 shows significant tumor growth inhibition in multiple human xenograft models at well-tolerated doses. Taken together, these data support the ongoing clinical investigation of XL765 for the treatment of cancer.

Kinase activity for PI3K isoforms was measured as the percentage of ATP consumed following the kinase reaction using luciferase–luciferin-coupled chemiluminescence as previously described (25), with ATP concentrations approximately equal to the K_m for each respective kinase. Kinase reactions were initiated by combining test compounds, ATP and kinase in a 20 μL volume. PI3Kα, PI3Kβ, PI3Kγ, and PI3Kδ (Upstate Biotechnology) final enzyme concentrations were 0.5, 8, 20, and 2 nmol/L, respectively. A similar assay format was used for DNAPK (DNA protein kinase; purchased from Promega) and VPS34 [PIK3C3; prepared at Exelixis as an N-terminal tagged full-length human fusion protein, which was expressed in insect cells using Baculovirus Expression Vector Systems (BEVS) and affinity purified using glutathione sepharose]. VPS34 assay buffer contained 20 mmol/L Tris-HCl, pH 7.5, 3.5 mmol/L MnCl₂, 100 mmol/L NaCl, 1 mmol/L DTT, and 0.01% cholamidopropyldimethylammonio propanesulfonate (CHAPS). Of note, 0.5 μL dimethyl sulfoxide (DMSO) containing varying concentrations of the test compound was mixed with 10 μL enzyme solution (2× concentration). Kinase reactions were initiated by the addition of 10 μL of liver phosphatidylinositol and ATP solution (2× concentration). Assay concentrations for VPS34, ATP, and phosphatidylinositol were 40 nmol/L, 1 μmol/L, and 5 μmol/L, respectively.

Cell lines were obtained from the American Type Culture Collection (ATCC) in 2001 to 2005 and maintained in culture conditions at 37°C under 5% CO₂. PC-3, MCF7, and A549 cells were maintained in Dulbecco's Modified Eagle Medium (DMEM; Cellgro 10-013-CV) containing 10% FBS (heat inactivated; Cellgro; 35-016-CV), 1% nonessential amino acids (NEAA; Cellgro; 30-002-CI), and 1% penicillin–streptomycin (Cellgro). U87-MG and MDA-MB-468 cells were maintained in Eagle's Minimum Essential Medium (EMEM)-Alpha (Cellgro; 10-022-CV), DMEM/F-12 (Cellgro, 15-090-CV), respectively, supplemented with 10% FBS, 2 mmol/L L-glutamine, 1% NEAA, and 1% penicillin–streptomycin. LS174T cells were maintained in MEM (GIBCO; 10370-021) containing 10% FBS, 2 mmol/l L-glutamine, 1 mmol/L sodium pyruvate, and 1% penicillin–streptomycin. Ramos cells were maintained in RPMI-1640 (Cellgro; 10-040-CV) containing 10% FBS (Cellgro) and 1% penicillin–streptomycin (Cellgro). OVCAR-3 cells were maintained in RPMI-1640 containing 20% FBS and 1% penicillin–streptomycin.

HEK 293 (ATCC) cells were grown in DMEM (Cellgro) containing 10% FBS (Cellgro), 1% NEAA (Cellgro), and 1% penicillin–streptomycin (Cellgro), and lysed in ice-cold lysis buffer containing 40 mmol/L HEPES pH 7.5, 120 mmol/L NaCl, 1 mmol/L EDTA, 10 mmol/L Na pyrophosphate, 10 mmol/L β-glycerophosphate, 50 mmol/L NaF, one tablet of protease inhibitors (Complete-Mini; EDTA-free; Roche), 0.3% CHAPS, and 1.5 mmol/L Na₃VO₄. mTORC1 was incubated with anti-mTOR antibody (N-19; Santa Cruz Biotechnology; sc-1549) 1.5 hours to overnight. The resulting immune-complexes were immobilized on immunoglobulin G (IgG) sepharose (GE Healthcare; 17-0618-01), washed sequentially three times with lysis buffer, once with wash buffer (50 mmol/L HEPES, pH 7.5, 40 mmol/L NaCl, and 2 mmol/L EDTA), and once with kinase buffer (25 mmol/L HEPES, pH 7.5, 50 mmol/L KCl, 20% glycerol, 10 mmol/L MgCl₂, 4 mmol/L MnCl₂, 1 mmol/L DTT). The immune-complexes (equivalent to 1 × 10⁶ cells) were preincubated at 30°C with XL765 or 0.1% DMSO for 10 minutes, and then subjected to kinase reaction for 30 minutes in a final volume of 20 μL (including 10 μL bed volume) containing kinase buffer, 25 μmol/L ATP, and l μg 4EBP1 (Exelixis). Kinase reactions were terminated by addition of 3.3 μL 4× sample buffer (Invitrogen; NP0007) containing 7% β-mercaptoethanol and analyzed by Western immunoblotting. Nitrocellulose membranes were incubated overnight at 4°C with 1/1,000 dilution of rabbit anti-mTOR (Upstate; 07-231) in 5% nonfat milk containing TBST (TBS/0.1% Tween-20) or with 1/500 dilution of rabbit anti-p4EBP1 [Cell Signaling Technology (CST), #9459] in 3% bovine serum albumin (BSA)/TBST, followed by incubation for 1 hours with a 1/5,000 dilution of secondary immunopure peroxidase–conjugated goat anti-rabbit IgG (H+L; Pierce; 31462) in 5% nonfat milk/TBST. Phospho-4EBP1 and mTOR were detected with Super Signal West Pico Substrate (Pierce; 34080). The p4EBP1 blot was subsequently stripped and reprobed with anti-4EBP1 antibody (1/1,000; CST#9452) with the total 4EBP1 signal detected as described above. Scans were analyzed using ImageQuant software. The DMSO control sample was used for normalization, and the IC₅₀ value for XL765 was determined using XLfit4 software.

HeLa (ATCC) cells were grown in suspension culture in EX-CELL HeLa media (Sigma; 14591C) and lysed as described above with minor modifications. The mTORC2 complex was incubated with anti-RICTOR antibody (Exelixis) for 2 hours and immune complexes (equivalent to 1 × 10⁷ cells) prepared as above with minor modifications. These were preincubated at 37°C with a test compound or 0.6% DMSO for 5 minutes, and then subjected to a kinase reaction for 8 minutes in a final volume of 33 μL (including 5 μL bed volume) containing kinase buffer, 50 μmol/L ATP, and 0.75 μg AKT1 [full-length human AKT1 with amino-terminal Hi tag was expressed in Sf9 cells using standard procedures, purified, then dephosphorylated with lambda protein phosphatase (New England Biolabs; P0753L) before repurification and use as substrate]. Kinase reactions were subsequently terminated and resolved as described above with minor modifications, then transferred onto polyvinylidene difluoride (PVDF) membranes at 50 V for 20 hours at 4°C. The membranes were blocked in 5% nonfat milk in TBST for 1 hour and incubated overnight at 4°C with 1/1,000 dilution of rabbit anti-pAKT (S473; CST# 4060) in 3% BSA/TBST. The membranes were washed three times in TBST and incubated for 1 hour with a 1/10,000 dilution of secondary goat anti-rabbit horseradish peroxidase (HRP) antibody (CST# 2125) in 5% nonfat milk/TBST. The IC₅₀ value for XL765 was determined as described above.

PC-3 (ATCC) and MCF7 (ATCC) cells were seeded at 2 × 10⁶ and 2.5 × 10⁶ cells, respectively, onto 10-cm dishes in culture medium and incubated at 37°C, 5% CO₂ for 24 hours. Growth medium was replaced with serum-free DMEM and cells were incubated for an additional 3 hours. Serial dilutions of test compounds in fresh serum-free medium were added to the cells in a final concentration of 0.3% DMSO (vehicle) and incubated for 23 minutes before recombinant human EGF stimulation (200 ng/mL; R&D Systems; 236-EG) for 2 minutes. After treatment, the medium was removed, and cellular material was precipitated with ice-cold 10% trichloroacetic acid (TCA) and collected by centrifugation. The pellet was washed with 3 mL of 5% TCA/1 mmol/L EDTA. Neutral lipids were extracted from the pellet with 3 mL of methanol:chloroform (2:1), and then the acidic lipids were extracted with 2.25 mL of methanol:chloroform:12 N HCl (80:40:1). The organic phase was separated from the aqueous phase by the addition of 0.75 mL of chloroform and 1.35 mL of 0.1 N HCl followed by centrifugation. The organic phase was then collected into a glass tube, dried under nitrogen gas, and resuspended by sonication in a water bath in 120 μL of the PIP₃ mass assay buffer (50 mmol/L HEPES, pH 7.4, 150 mmol/L NaCl, and 1.5% sodium cholate).

Assays were conducted in 96-well plates (PerkinElmer; L2251692) by incubating 50 μL of the lipid extract with 50 μL of the sensor complex in detection buffer (10 mmol/L Tris-HCl, pH 7.2, 150 mmol/L NaCl, 7.5 mmol/L EDTA, 0.1% Tween-20, and 1 mmol/L DTT) at ambient temperature in the dark for 2 hours, and plates were read using an AlphaQuest reader (PerkinElmer). The sensor complex contained 15 μL of 100 nmol/L biotinylated PIP₃ (Echelon; C-39B6), 15 μL of 100 nmol/L glutathione S-transferase (GST)–tagged GRP1 pleckstrin homology (Echelon; G-1200), and 20 μL of a mixture of donor and acceptor AlphaScreen beads (GST detection kit; PerkinElmer; 6760603c). The PIP₃ mass present was estimated by comparison with standard curves constructed by addition of known amounts of diC8 PI(3,4,5)P3 standard (Echelon; P-3908) to the sensor complex.

pS6 ELISA was performed as previously described (26) with minor modifications. The pAKT ELISA assay was performed as follows: PC-3 (ATCC) cells were seeded at 1.5 × 10⁵ cells per well in 6-well plates (NUNC; 140685) in growth medium then incubated at 37°C, 5% CO₂ for 72 hours, and the growth medium was replaced with serum-free DMEM. Serial dilutions of the test compound in 0.3% DMSO (vehicle) were added to the cells and incubated for 2 hours and 50 minutes. Cells were then stimulated with 100 ng/mL EGF (R&D Systems; 236-EG) for 10 minutes. Cells were washed once with ice-cold PBS, harvested by briefly shaking in 100 μL of TENN lysis buffer (20 mmol/L Tris-HCl, pH 7.5, 1 mmol/L EDTA pH 8.0, 0.5% NP40, 150 mmol/L NaCl) with protease and phosphatase inhibitors (1 mmol/L PMSF (phenylmethylsulfonylfluoride), 1 μg/mL leupeptin, 1 μg/mL aprotinin, 1 μg/mL pepstatin, 1 mmol/L EDTA, 1 mmol/L NaF, 20 mmol/L β-glycerophosphate, 1 mmol/L Na-orthovanadate, and 5 mmol/L p-nitrophenyl phosphate), and transferred to 96-well plates. Cells were lysed on ice for 20 minutes by pipetting up and down 10 times, and the supernatants were collected by centrifugation. ELISA assays for pAKT^T308 and total AKT were performed with the AKT[pT308] ELISA Kit (BioSource International; KHO0201) and AKT ELISA Kit (BioSource International; KHO0101). IC₅₀ values were determined on the basis of the ratio of pAKT to total AKT signal in lysates from compound-treated cells, normalized to lysates from DMSO-treated controls.

PC-3, MCF7, A549, U87-MG, LS174T, MDA-MB-468, and OVCAR-3 cells were seeded at densities of 2 × 10⁶, 2 × 10⁶, 1.2 × 10⁶, 2 × 10⁶, 2 × 10⁶, 1.2 × 10⁶, and 1 × 10⁶ cells, respectively, onto 10-cm dishes in their respective culture medium and incubated at 37°C, 5% CO₂ for 20 to 47 hours. The medium was replaced with test compounds dissolved in the same media containing 0.3% DMSO, and the cells were incubated for 3 hours. For growth factor treatment, the medium was replaced with test compounds dissolved in serum-free DMEM containing 0.3% DMSO. After incubation for 3 hours, cells were stimulated with 100 ng/mL of EGF (R&D Systems; 236-EG) for 10 minutes. Cells were washed with ice-cold PBS, and directly lysed with cell lysis buffer (BioSource International; FNN0011) containing protease inhibitors (Complete-Mini; EDTA-free; Roche; 11836170001; aminoethylbenzenesulfonyl fluoride; Sigma; A8456). Protein lysates were analyzed by Western immunoblotting. PVDF membranes (Invitrogen) were incubated overnight at 4°C with primary antibodies at the indicated concentrations in 3% BSA/TBST buffer, followed by incubation for 1 hour with a 1/10,000 dilution of secondary goat anti-rabbit HRP antibody (CST#7074) or a 1/3,000 dilution of secondary goat anti-mouse HRP antibody (Amersham, NXA931) in 5% nonfat milk/TBST. Signals were detected using ECL-plus (Amersham; RPN2132) and scanned using a Typhoon 9400 scanner (Molecular Devices). For total protein readouts, stripped membranes were incubated with the respective primary antibodies, with signal detection as described above. Scans were analyzed using ImageQuant software. Phospho signals were normalized to the corresponding total protein signals, the percentage of inhibition relative to DMSO control was determined, and IC₅₀ values were calculated using XLfit4 software. The following antibodies were used in Western Immunoblot analysis: pAKT (T308; CST#4056; 1/500 dilution), pAKT (S473; BioSource International; 44-622G; 1/1,000), AKT (BioSource International; 44-607G; 1/1,000 dilution), pp70S6K (T389; CST #9234; 1/1,000 dilution), p70S6K (Bethyl; A300-510A; 1/2,500 dilution), pS6 (S240/244; CST #2215; 1/2,000 dilution), S6 (CST #2217; 1/2,000 dilution), pPRAS40 (T246; BioSource International; 44-1100G; 1/2,000 dilution), PRAS40 (BioSource International AHO1031; 1/1,000 dilution), pGSK3β (S9; CST #9336; 1/1,000 dilution), GSK3β (CST #9315; 1/1,000 dilution), p4EBP1 (T37/46; CST #9459; 1/1,000 dilution), 4EBP1 (CST #9452; 1/1,000 dilution), cyclin D1 (EMD Biosciences; CC12; 1/1,000 dilution), pERK (Y204; Santa Cruz Biotechnology; sc-7383, 1/1,000 dilution), and ERK (extracellular signal–regulated kinase; CST #9102; 1/1,000 dilution).

Ramos (ATCC) cells were seeded at a density of 0.3 × 10⁶ cells/mL in growth medium. The next day, cells were centrifuged, washed with serum-free medium supplemented with 1% BSA (tissue culture tested; Sigma; A4919), resuspended in RPMI supplemented with 1% BSA, and incubated for 20 hours. The serum-starved cells were centrifuged and resuspended in 5 mL of the saved media at 1 × 10⁶ cells/mL. Cells were treated with test compound administered at a final DMSO concentration of 0.3%. For the nutrient-depletion control, the serum-starved cells were washed once with PBS, and resuspended in 5 mL of PBS with 0.3% DMSO. After incubation for 2 hours, cells were washed once in PBS and lysed in 150 μL of BioSource cell lysis buffer containing protease inhibitors (Complete-Mini; EDTA-free; Roche; AEBSF; Sigma). Lysates were analyzed by Western immunoblotting. PVDF membranes were incubated overnight at 4°C with 1/500 dilution of rabbit anti-pmTOR (CST#2974), with 1/1,000 dilution of rabbit anti-pp70S6K (CST#9234), or with 1/1,000 dilution of rabbit anti–p4E-BP1 (CST#9459) in 3% BSA/TBST, followed by incubation for 1 hours with a 1/10,000 dilution of secondary goat anti-rabbit HRP antibody (CST#7074) in 5% nonfat milk/TBST and signals detected and analyzed as described above. For detection of total p70S6K and total 4E-BP1, stripped membranes were probed with rabbit antitotal p70S6K (1/2,500, Bethyl; A300-510A) and with rabbit antitotal p4E-BP1 (1/1,000; CST#9452) with signals detected and analyzed as described above.

Cellular proliferation was assessed as previously described (27) using the Cell Proliferation ELISA, Bromodeoxyuridine (BrdUrd) Chemiluminescence Kit (Roche; Applied Science). Cytotoxicity was assessed using the ATP Bioluminescence Assay as follows: PC-3, MCF7, A549, LS174T, MDA-MB-468, U87-MG, and OVCAR-3 cells were plated at densities of 7 × 10³, 1.5 × 10⁴, 6 × 10³, 7 × 10³, 7 × 10³, 6 × 10³, 1.5 × 10⁴ cells per well, respectively, onto 96-well microtiter plates (Corning; 3904) in culture medium, incubated at 37°C, 5% CO₂ for 18 hours, and then treated with a serial dilution of compound in medium containing a final concentration of 0.3% DMSO. Triplicate wells were used for each compound concentration. Control wells received 0.3% DMSO in media. Cultures were incubated at 37°C, 5% CO₂ for an additional 24 hours and cells were then assayed for viability using the ViaLight HS Kit (Cambrex; LT07-111).

PC-3, MCF7, A549, LS174T, MDA MB 468, U87 MG, and OVCAR-3 cells were plated at densities of 5 × 10³, 1.2 × 10⁴, 5 × 10³, 6 × 10³, 6 × 10³, 5 × 10³, and 1.2 × 10⁴ cells per well, respectively, onto 96-well microtiter plates (Corning; 3904), in culture medium at 37°C, 5% CO₂ for 18 hours, and then treated with a serial dilution of compound in medium containing a final concentration of 0.3% DMSO. Triplicate wells were used for each compound concentration. Positive control wells received 5 to 30,000 nmol/L adriamycin (MCF7, A549, and LS174T) or 5 to 30,000 nmol/L camptothecin (PC-3, U87-MG, OVCAR-3, and MDA-MB-468) and negative control wells received 0.3% DMSO in media. Background wells contained no cells and 0.3% DMSO in media. Following incubation at 37°C, 5% CO₂ for an additional 48 hours, apoptosis was assessed using the Apo-ONE Homogeneous Caspase-3/7 Assay Kit (Promega; G7791). EC₅₀ values were calculated on the basis of the fluorescence of compound-treated wells compared with that of the corresponding positive control.

Soft agar (60 μL/well of 0.75%; BD Biosciences) was plated in a 96-well black plate (Nalge Nunc International) and allowed to solidify at 37°C for 20 minutes. A total of 4.8 × 10³ PC-3 or MCF7 cells in 100 μL of media containing 0.375% agar, FBS (15% for PC-3 and 20% for MCF7), and 1× concentrations of serial dilutions of XL765 were layered over the base agar. After 10 minutes, 60 μL of DMEM (GIBCO) containing FBS and 2-fold concentrated test compounds were added over the cell layer. Following equilibration of the media and the test compound, the final compound concentration was presumed to be 1× in all three layers. The cultures were incubated for 14 days at 37°C, 5% CO₂. At day 7, 50 μL of fresh media containing 10% FBS without compound was added to keep the cultures from drying out. At the completion of the incubation, the media in the top layer were removed, and 40 μL of media containing 50% Alamar Blue (BioSource International; DAL1025) were added to each well followed by incubation at 37°C, 5% CO₂ for 4 hours and subsequent fluorescent detection.

The hepatocyte growth factor (HGF)–induced chemotaxis assay was performed as previously described (27).

PC-3, MCF7, and B16F10 (ATCC) cells were mixed with a series of diluted compounds in serum-free DMEM (Gibco), DMEM containing 0.2% FBS, and serum-free EBM-2 medium (Clonetics), respectively. Control wells received media with 0.25% DMSO alone. A total of 5 × 10³ cells were plated in each well of a 96-well plate and incubated at 37°C, 5% CO₂ for 18 hours (PC-3) or 24 hours (B16F10 and HMVEC-L). At the end of the incubation, cell viability was determined using Alamar Blue solution (BioSource International).

Tumors were collected at the indicated time points and tumor lysates were prepared as previously described (25). Pooled lysates were analyzed by Western immunoblotting. PVDF membranes were incubated overnight at 4°C with primary antibodies to the respective phosphoepitopes at the indicated concentrations in 3% BSA/TBST buffer. The membranes were then incubated for 1 hour with a 1/10,000 dilution of secondary goat anti-rabbit HRP antibody (CST#7074) in 5% nonfat milk/TBST. Signals were detected using ECL-plus (Amersham; RPN2132) and scanned using Typhoon (Molecular Devices). To determine total protein levels, membranes were stripped and incubated with the indicated primary antibodies specific for the respective total proteins. The same procedure described above was followed to detect the total protein signal. Scans were analyzed using Image Quant software. The percentage of inhibition was determined by normalizing the phosphoepitope signals to the total protein signals and then calculating the percentage of inhibition compared with vehicle control groups. The following antibodies were used in Western immunoblot analysis of tumor extracts: pAKT (T308; CST#4056; 1/500 dilution), pAKT (S473; CST #9271; 1/1,000), AKT (CST #9272; 1/1,500 dilution), pp70S6K (T389; CST #9234; 1/500 dilution), p70S6K (Bethyl; A300-510A; 1/2,000 dilution), pS6 (S240/244; CST #2215; 1/2,000 dilution), and S6 (CST #2217; 1/1,500 dilution).

In vivo efficacy studies were performed in athymic nude mice purchased from Taconic and housed according to the Exelixis Institutional Animal Care and Use Committee guidelines. Tumor cells were cultured in vitro in DMEM (Mediatech) supplemented with 10% FBS (20% for PC-3 and OVCAR-3 cells), penicillin—streptomycin, and nonessential amino acids at 37°C in a humidified 5% CO₂ atmosphere. On day 0, cells were harvested by brief trypsinization, and 1 to 5 × 10⁶ cells in 0.1 mL ice-cold Hanks Balanced Salt Solution were implanted subcutaneously (OVCAR-3) or intradermally (MCF7 and U-87 MG) into the hind flank of female athymic nude mice. In the case of the MCF7 model, an estrogen pellet (IRA) was implanted subcutaneously at the nape of neck at the time of tumor cell implantation. A total of 3 × 10⁶ PC-3 cells were similarly harvested and implanted subcutaneously into the hind-flank of 5- to 8-week-old male nude mice. Tumor growth was monitored weekly with calipers until staging and dose initiation. During the dosing period, body and tumor weights were assessed as previously described (27). XL765 was formulated in sterile water/10 mmol/L HCl or water and administered at the indicated doses and regimens by oral gavage at a dose volume of 10 mL/kg.

After euthanasia, tumors from animals administered XL765 and/or other agents were excised and fixed in zinc fixative (BD Pharmingen) for 24 to 48 hours before being processed into paraffin blocks. Of note, 5-μm-thin sections were cut serially to represent the largest possible surface for each tumor and stained using standard immunohistochemical methods to detect Ki67 nuclear antigen (LabVision) and CD31-positive tumor vessels (BD Pharmingen). CD31 was detected by biotinylated secondary antibody followed by the avidin–biotin–peroxidase complex (BD Pharmingen). Ki67 was detected by Envision⁺ anti-rabbit peroxidase complexed polymer (DAKO). Sections were counterstained with hematoxylin. The tumor mean vessel density and Ki67 index in tumor sections were quantified using the ACIS automatic cellular imaging system (Clarient Inc.). The mean number of tumor vessels per mm² was determined by analyzing eight to 15 fields across the total tumor section. The percentage of Ki67-positive tumor cells was determined by sampling multiple representative fields of equal size across the total viable tumor area of each section and dividing the number of Ki67-positive cells by the total number of cells identified per field. Apoptosis was assessed by TUNEL as previously described (27). The results for each immunohistochemical readout were averaged for each tumor section, followed by averaging the results for each treatment group (n = 9–10). Statistical analyses were performed using the standard two-tailed t test with Bonferroni adjustment for multiple comparisons against a single control group.

XL765 (Fig. 1A) was identified following optimization of a pyridopyrimidinone scaffold for in vivo PI3K/mTOR pathway inhibition and drug-like properties. In assays performed using purified proteins in a luciferase-coupled chemiluminescence format, XL765 displayed potent inhibitory activity against class I PI3K isoforms p110α, p110β, p110δ, and p120γ, with IC₅₀ values of 39, 110, 43, and 9 nmol/L, respectively (Table 1). The IC₅₀ value for inhibition of PI3Kα by XL765 was determined at various concentrations of ATP, revealing XL765 to be an ATP-competitive inhibitor with an equilibrium inhibition constant (K_I) value of 13 nmol/L.

XL765 also inhibited mTOR (IC₅₀ values of 160 and 910 nmol/L for mTORC1 and mTORC2, respectively) in an immune-complex kinase assay and the PI3K-related kinase DNAPK (IC₅₀ value of 150 nmol/L). In contrast, XL765 had relatively weak inhibitory activity toward the class III PI3K vacuolar sorting protein 34 (VPS34; IC₅₀ value of ∼9.1 μmol/L). XL765 was also profiled against a panel of approximately130 protein kinases; no cross-reactivity was observed at concentrations below 1.5 μmol/L (Supplementary Table S1). All assays were performed at ATP concentrations approximately equal to the Michaelis constant (K_M) values of the respective enzymes.

MCF7 human mammary carcinoma cells and PC-3 human prostate adenocarcinoma cells were selected for the initial assessment of the effect of XL765 on signaling downstream of PTEN/PI3K because they each have a prevalent genetic lesion that activates the PI3K pathway. MCF7 cells carry a heterozygous E545K-activating mutation in the p110α subunit of PI3K and PC-3 cells carry a homozygous deletion of exons 3 to 9 of the PTEN tumor suppressor gene. PIP₃ is the product of a class I PI3Ks acting on the physiologic substrate PIP₂. Hence, PIP₃ levels serve as a direct assessment of PI3K activity. Consistent with its inhibitory activity against purified PI3K proteins, XL765 inhibited EGF-induced PIP₃ production in PC-3 and MCF7 cells with IC₅₀ values of 290 and 170 nmol/L, respectively (Table 2). The ability of XL765 to inhibit phosphorylation of key signaling proteins downstream of PI3K was examined by assessing its effects on EGF-stimulated phosphorylation of AKT and on nonstimulated phosphorylation of S6 in PC-3 cells by cell-based ELISA. XL765 inhibited these activities with IC₅₀ values of 250 and 120 nmol/L, respectively (Table 2).

The effects of XL765 on the PI3K signaling pathway were then examined by Western immunoblot analysis in PC-3 and MCF7 cells (see Fig. 1B and Supplementary Fig. S1). The results were consistent in both cell lines. XL765 inhibits AKT phosphorylation at both activation sites (T308 and S473) at concentrations consistent with the IC₅₀ values determined by ELISA. The T308 phosphorylation site on AKT is a substrate for PDK1 (1), whereas the S473 site is a substrate for mTORC2 (6). Inhibition of AKT substrate phosphorylation (PRAS40 and GSK3β) and inhibition of phosphorylation events downstream of mTOR (p70S6K, S6, and 4EBP1 phosphorylation) were also evident. XL765 induces a decrease in the levels of cyclin D1 protein, consistent with increased GSK3β activity as a result of inhibition of AKT leading to GSK3β-mediated phosphorylation and subsequent degradation of cyclin D1 (Fig. 1B). Overall, a similar range of compound concentrations was required to inhibit PI3K proximal phosphorylation events (AKT T308 phosphorylation) and phosphorylation events downstream of mTORC1 and mTORC2 (p70S6K phosphorylation and AKT S473 phosphorylation).

The control compound ZSTK474 (an inhibitor of PI3K; Reference 28) at 10 μmol/L robustly decreased the levels of all the phospho readouts assessed. The TORC1 inhibitor rapamycin at 0.1 μmol/L did not inhibit the phosphorylation of AKT or its direct substrates PRAS40 and GSK3β, but in fact seemed to stimulate phosphorylation of AKT. This is consistent with relief of p70S6K-dependent negative feedback of PI3K (see Introduction). As expected, rapamycin significantly decreased p-p70S6K and pS6 levels, consistent with its well-characterized ability to inhibit mTORC1. None of the compounds had significant effects on ERK1/2 phosphorylation, consistent with biochemical profiling data. XL765 was further profiled in additional cell lines bearing a variety of genetic lesions that activate/modulate the PI3K pathway. These were OVCAR-3 (PIK3CA amplification), U87-MG (PTEN deletion), A549 (KRAS mutation, loss-of-function mutation in the mTOR-directed tumor suppressor gene LKB-1), MDA-MB-468 (PTEN deletion), and LS174T (PIK3CA and KRAS mutations) cells. XL765 demonstrated consistent activity in these cell lines with no marked differences in sensitivity being evident (Supplementary Figs. S2 and S3).

To directly assess the impact of XL765 on mTOR in cells, we used an approach that relies on the ability of nutritional signals to activate mTOR independent of PI3K activity. Regulation of mTOR signaling by nutrient availability is predominant in certain transformed B cells (29), and we used the Burkitt lymphoma–derived cell line Ramos as a system to study the effect of XL765 on nutrient-dependent mTOR activity (Fig. 1C). Cells were starved in serum-free media for 20 hours, and then treated with compounds, DMSO, or serum- and nutrient-free PBS for 2 hours. Cell lysates were prepared and analyzed by gel electrophoresis and Western immunoblotting with anti-pmTOR, anti–p-p70S6K, and anti-p4EBP1 antibodies. Cells incubated in PBS show very low levels of phosphorylation of p70S6K, 4EBP1, or the mTOR autophosphorylation site S2481, consistent with low mTOR activity. Incubation of cells in serum-free, nutrient containing media results in a robust upregulation of mTOR-dependent phosphorylation events. ZSTK474 and PI-103 (30) at 10 μmol/L inhibit these phosphorylation events, consistent with their ability to directly inhibit mTOR kinase activity in addition to PI3K activity. Rapamycin treatment at 0.1 μmol/L resulted in little if any decrease in mTOR autophosphorylation, but profoundly inhibited p70S6K phosphorylation consistent with selective inhibition of mTORC1. XL765 inhibited nutrient-dependent phosphorylation at all sites, with IC₅₀ values of 160, 340, and 3,000 nmol/L, for mTOR S2481, p70S6K, and 4EBP1 phosphorylation, respectively. These results are consistent with the mTOR kinase assay results presented above, and provide further evidence that XL765 is a direct inhibitor of mTOR kinase activity.

In MCF7 and PC-3 cells, XL765 inhibits proliferation (monitored by BrdUrd incorporation) with IC₅₀ values of 1,070 and 1,840 nmol/L, respectively. When tested in a broad panel of tumor cell lines with diverse origins and genetic backgrounds, XL765 was found to inhibit proliferation with a wide range of IC₅₀ values (200 to >30,000 nmol/L; Fig. 2; Supplementary Table S2). A breakdown of sensitivity by genotype suggested that PIK3CA-mutant cell lines tended to be relatively sensitive to XL765, whereas RAS- or BRAF-mutant cell lines tended to be less sensitive. Interestingly, several RAS-mutant cell lines were relatively insensitive to XL765 despite of their also harboring PIK3CA mutations (Fig. 2; Supplementary Table S2). Cell lines with loss of PTEN showed a range of sensitivities, with some (e.g., the prostate carcinoma lines ZR75-1, LNCap, and PC-3) being sensitive and others (e.g., the glioblastoma cell lines U251, U373) being refractory.

Anchorage-independent growth in soft agar is considered the most stringent assay for detecting malignant transformation of cells. To further characterize the effects of XL765 on tumor cell growth, an assay monitoring the anchorage-independent growth of PC-3 and MCF7 cells in soft agar over a 14-day period was used. XL765 inhibits colony growth with an IC₅₀ value of 270 nmol/L in PC-3 cells and 230 nmol/L in MCF7 cells. These IC₅₀ values are significantly lower than those required to inhibit growth of the cells in a monolayer, perhaps indicating an increased reliance on PI3K pathway signaling for growth in three-dimensions.

To rule out direct cytotoxic effects of XL765 on tumor cells, its effects on cell viability were determined by bioluminescent measurement of cellular ATP. XL765 did not reduce ATP levels in cells when incubated for 24 hours, indicating a lack of acute cytotoxicity (Supplementary Table S2, footnote). Induction of cytoplasmic caspases 3 and 7 was examined as an indication of apoptosis induction. XL765 did not affect the activity of these caspases at the doses and time point tested (Supplementary Table S2, footnote). In MCF7 cells, the antiproliferative effects of XL765 were associated with a specific block in the G₁ phase of the cell cycle and an increase of sub-G₁ cell population (data not shown). Therefore, at least in MCF7 cells, XL765 exhibits antiproliferative effects predominantly via blockade of the cell cycle rather than through cytotoxic or apoptotic effects.

One of the hallmarks of aggressive tumor cells is the ability to migrate in response to chemotactic stimuli and to invade surrounding tissue. HGF is one of the key stimulators of these behaviors, and cell lines expressing high levels of the HGF receptor MET, are highly invasive and metastatic in vivo. Because PI3K resides in the MET signaling pathway, the ability of XL765 to inhibit HGF-stimulated migration was tested in vitro. Murine B16 melanoma cells express high levels of Met, which becomes highly phosphorylated when the cells are treated with HGF. In 10% serum, B16 cells plated in the top well of a Transwell chamber containing a barrier with 0.8-μm pores show very little ability to migrate to the lower chamber side. Addition of HGF to the lower Transwell chamber greatly increases migration through the barrier over a 24 hours period. XL765 blocks this effect with an IC₅₀ value of 601 nmol/L (Supplementary Fig. S4). The cytotoxicity IC₅₀ value of XL765 in B16 cells is 7,300 nmol/L, 12-fold higher than the IC₅₀ value for inhibition of migration. Therefore, inhibition of melanoma cell migration by XL765 is unlikely to be due to cytotoxicity.

Lysates derived from MCF7 xenograft tumors intradermally implanted into athymic nude mice contain high levels of constitutively phosphorylated AKT, p70S6K, and S6 proteins. The ability of XL765 to inhibit endogenous phosphorylation of AKT, p70S6K, and S6 was examined following a single oral dose of 10, 30, 100, or 300 mg/kg. The tumors were harvested 4, 24, or 48 hours after dose and homogenized in lysis buffer. Tumor lysates from each animal (n = 4) were then pooled for each group and analyzed for levels of total and phosphorylated AKT, p70S6K, and S6 by Western immunoblotting (Fig. 3A).

Oral administration of XL765 causes a dose-dependent decrease of phosphorylation of AKT, p70S6K, and S6 in the tumors, reaching a maximum of 84% inhibition of S6 phosphorylation at 30 mg/kg at 4 hours. The dose–response relationships (not shown) derived from the 4 hours time point predict 50% inhibition of AKT, p70S6K, and S6 phosphorylation to occur at doses of 19 mg/kg (pAKT^T308 and pAKT^S473), 51 mg/kg (p-p70S6K), and 18 mg/kg (pS6). Inhibition of AKT, p70S6K, and S6 phosphorylation in MCF7 tumors following a 30 mg/kg dose of XL765 was maximal at 4 hours, reaching 61% to 84%; however, the level of inhibition decreased to 0% to 42% by 24 hours, and minimal or no inhibition was evident by 48 hours (see Fig. 3A). Following a 100 mg/kg dose of XL765, inhibition was also maximal at 4 hours (52%–75%). However, in contrast with the 30 mg/kg dose, inhibition at 24 hours (48%–71%) was almost comparable with that seen at 4 hours. Partial inhibition of some phosphoepitopes persisted through 48 hours (Fig. 3A).

Similarly, administration of XL765 caused a dose-dependent decrease of phosphorylation of AKT, p70S6K, and S6 in PC-3 tumors in vivo, reaching a maximum of 93% inhibition of AKT phosphorylation at 300 mg/kg at 4 hours after dose (Fig. 3B). The dose–response relationships (not shown) derived from the 4 hours time point predict 50% inhibition of AKT, p70S6K, and S6 phosphorylation to occur at doses of 15 mg/kg (pAKT^T308), 13 mg/kg (pAKT^S473), 59 mg/kg (p-p70S6K), and 48 mg/kg (pS6). Consistent with the MCF7 data, for the 100 mg/kg dose inhibition (42%–60%) persisted through 24 hours after dose. In both studies, blood was collected at the same time tumor tissue was harvested and plasma concentrations of XL765 were assessed (Supplementary Table S3). On the basis of these data, the plasma concentrations associated with inhibition of phosphorylation of AKT, p70S6K, and S6 by 50% in these tumor models ranged from approximately 3 to 9 μmol/L. Hence, XL765 exhibited comparable pharmacodynamic activity in PIK3CA-mutant MCF7 and PTEN-deficient PC-3 xenograft tumor models.

Multiple tumor models were used to explore the efficacy and potency of repeat-dose oral administration of XL765 with regard to tumor growth inhibition in vivo. In addition to the previously described MCF7 and PC-3 models, the antitumor efficacy of XL765 was evaluated in xenograft models, including OVCAR-3 (human ovarian xenograft tumor model exhibiting PIK3CA amplification), U-87 MG (human glioblastoma xenograft tumor model harboring a deletion at codon 54 in the gene encoding PTEN, resulting in a frameshift), A549 [human non–small lung cancer cell (NSCLC) xenograft tumor model harboring a homozygous-activating mutation in KRAS and a homozygous loss-of-function mutation in LKB1], and Calu-6 (human NSCLC xenograft tumor model harboring an activating mutation in KRAS).

XL765 administration results in significant antitumor efficacy in vivo in all of these models (Fig. 4) at doses that proved well tolerated as assessed by daily monitoring of mouse weights (Supplementary Fig. S5; no or minimal impact on body weights compared with vehicle control). The most efficacious schedules were 30 mg/kg twice a day and 100 mg/kg every 2 days, which suggests that sustained pathway inhibition is required for maximal effect on tumor growth. These schedules generally resulted in stasis of tumor growth, except in the PC-3 model and the Calu-6 KRAS-mutant NSCLC model, in which tumors continued to grow although at a reduced rate. Immunohistochemical analyses of MCF7, PC-3, and A549 tumors collected at the end of the dosing period revealed significant, dose-dependent decreases in staining for Ki67, a marker of cell proliferation. Moreover, XL765 administration was associated with increased tumor cell apoptosis in MCF7 and A549 tumors and modestly decreased tumor vascularization in MCF7, PC-3, and A549 tumors (Table 3). Thus, inhibition of PI3K and mTOR by XL765 results in antiproliferative, proapoptotic and antivascular effects in xenograft tumors. Plasma concentrations at the end of these efficacy studies were similar to those seen following single-dose administration. For example, in the MCF7 efficacy study, average plasma concentrations for the 30 mg/kg dose administered once daily were 8.8, 5.3 μmol/L, and below the limit of detection at the 1, 4, and 24 hours time points, respectively (n = 3/time point).

The antitumor activity observed in the subcutaneous U-87 MG glioblastoma xenograft model prompted us to examine the pharmacodynamic activity of XL765 in the mouse brain as a measure of whether the compound could effectively cross the blood–brain barrier. Lysates from brains of nontumor-bearing mice show significant PI3K pathway activity as judged by levels of pAKT and pS6. Four hours following a single oral dose of 30 or 100 mg/kg XL765, pAKT and pS6 levels are substantially reduced, demonstrating that XL765 can cross the blood–brain barrier and inhibit the PI3K pathway (Fig. 4).

Previous experience with highly selective inhibitors of signal transduction pathways has revealed the existence of unanticipated regulatory mechanisms that can act to limit the efficacy of pathway inhibition. An example is the upregulation of AKT phosphorylation that occurs as the result of relief of a negative feedback loop following inhibition of mTORC1 by the rapamycin class of mTOR inhibitors. Similarly, selective B-RAF inhibitors can trigger a “paradoxical” activation of C-RAF in the context of activated RAS as a result of allosteric effects on B-RAF/C-RAF dimers. Inhibition at a single node in a pathway also allows for the development of resistance via pathway activation downstream of the point of intervention. Combining multiple inhibitors that affect the same pathway, or developing a single compound that inhibits multiple members of a pathway is, therefore, an attractive approach to limit or circumvent these issues. For example, emerging data suggest that combinations of B-RAF and MEK inhibitors are superior to either agent alone for the treatment of B-RAF–mutant metastatic melanoma.

The PI3K/mTOR pathway is one of the most frequently activated signaling pathways in human cancer, in part, due to the mutation, amplification, or deletion of key pathway regulatory components. Activation of the pathway promotes tumor cell proliferation, survival, and resistance to anticancer therapies. As a result, extensive efforts are being devoted to identifying and developing small-molecule inhibitors that affect different nodes of the pathway. Activation of the pathway is subject to regulation at multiple points and by a wide variety of signals, including growth factors, cellular energy levels, nutritional status, and oxygenation. We, therefore, elected to optimize a compound that would inhibit two key nodes in the pathway, PI3K and mTOR, with the aim of maximizing pathway blockade in the context of multiple genetic backgrounds and under a variety of environmental conditions.

XL765 (SAR245409) inhibits both class I PI3Ks as well as mTOR. Although XL765 seems to be more potent against PI3Kα than against PI3Kβ when assessed by biochemical assays using purified kinases, in cellular assays XL765 shows comparable activity versus PI3K pathway signaling in MCF7 breast (PIK3CA-mutant) and PC-3 prostate (PTEN-deleted) tumor cells. PI3Kβ is considered to be the major driver of dysregulated PI3K pathway activity associated with PTEN deficiency (31).

Moreover, XL765 showed comparably robust and persistent pharmacodynamic activity against these cell lines when they were grown as xenograft tumors in mice. These data demonstrate that XL765 exhibits functionally comparable activity against PI3Kα and PI3Kβ in cultured cells and xenograft tumors.

In biochemical assays XL765 was generally less potent against mTOR than against class I PI3K isoforms. In tumor cells, however, XL765 inhibits mTOR-dependent phosphorylation events and PI3K-independent, nutrient-stimulated mTOR activity with a potency comparable with that demonstrated for PI3K-dependent signaling, suggesting potential for concerted PI3K/mTOR inhibition in cellular and in vivo models.

Our survey of the effects on XL765 on multiple phospho-epitopes in the PI3K signaling pathway in six tumor cell lines with differing genetic backgrounds revealed consistent inhibition downstream of both PI3K and mTOR. We saw no evidence for feedback upregulation within the pathway or with respect to ERK phosphorylation. However, it is important to note that our data are not exhaustive with respect to either genotype or phosphorylation sites surveyed, and are limited to a single time point. The profile of XL765 is clearly differentiated from that of rapamycin, which in all cell lines tested is a potent inhibitor of mTORC1-dependent p70S6K and S6 phosphorylation, but in some cell lines augmented PI3K activity as assessed by AKT phosphorylation, consistent with previous reports.

XL765 exhibits a wide range of antiproliferative activity against tumor cells grown as monolayers. In MCF7 cells, these effects were associated with a G₁ arrest but not with acute cytotoxicity or induction of apoptosis. When sensitivity to XL765 is examined in relation to genotype, there is a trend suggesting enhanced sensitivity of cells exhibiting PIK3CA-activating mutations, consistent with similar observations previously reported for the PI3K inhibitors GDC-0941 and CH5132799 (32, 33). Likewise, the relative insensitivity of RAS-mutant cell lines, regardless of PIK3CA status, to inhibition of proliferation by XL765 is consistent with preclinical observations with other PI3K pathway–targeting agents (23). However, we note that a number of K-RAS–mutant lines such as the A549 NSCLC line were quite sensitive to XL765. These cells have a concomitant deletion of the LKB1 gene, which may serve to sensitize them to inhibition of mTOR. Likewise, PTEN-deleted cell lines had a wide range of sensitivities to XL765 with some being very sensitive and some being refractory. The basis for these differences is not understood, but presumably reflects varying degrees of dependence on PI3K pathway signaling for proliferation as a result of alterations in other pathways that affect growth.

In multiple xenograft tumor models, oral administration of XL765 resulted in substantial tumor growth inhibition at well-tolerated doses. The most efficacious schedules were 30 mg/kg twice a day or 100 mg/kg every 2 days, suggesting that efficacy is associated with more continuous inhibition of the pathway. These models encompass multiple genetic lesions activating the PI3K pathway, specifically a PIK3CA E545K mutation (MCF7), PIK3CA amplification (OVCAR-3), PTEN deletion (PC-3 and U-87 MG), KRAS mutation (A549 and Calu-6), and LKB1 mutation (A549). The fact that efficacy was observed in all these models suggests that XL765 may have broad utility in tumors with activation of the PI3K pathway. On the basis of the immunohistochemical/immunofluorescence analyses conducted on tumor xenografts following repeat dosing of XL765, antitumor efficacy was associated with a combination of antiproliferative and proapoptotic effects, with a modest impact on tumor angiogenesis. These proapoptotic effects in vivo, which are not evident on cultured tumor cells, likely reflect targeting of the tumor microenviroment in addition to tumor cells themselves, which is consistent with the antiangiogenic effects evident.

In the majority of the xenograft models, complete or near complete inhibition of tumor growth (but not regression) was observed, with the exception of the PC-3 and Calu-6 models, which were relatively resistant. Overall, our data are not extensive enough to determine whether in vitro sensitivity is generally predictive of efficacy in xenograft models. It is also not yet clear whether the presence of PIK3CA mutations or PTEN deficiency will be predictive of greater clinical responsiveness to PI3K pathway inhibitors in general, although an analysis based on combining the results of multiple early-stage trials suggested that PIK3CA H1047R mutations are associated with response (34). Intensive molecular profiling of tumors in the ongoing XL765 clinical studies is being performed to further explore this question.

Because we observed significant antitumor efficacy in the U-87 MG glioblastoma model, we assessed the pharmacodynamic activity of XL765 in mouse brain. At the same doses associated with efficacy in the subcutaneous xenograft model, XL765 effectively inhibited PI3K pathway signaling in the brain, supporting its potential utility for the treatment of central nervous system malignancies. Consistent with this observation, XL765 has demonstrated significant efficacy in an orthotopic glioblastoma xenograft model, both as a single agent and in combination with temozolomide (35). In addition, recent data from a clinical trial in which XL765 was administered to glioblastoma patients before surgical removal of recurring lesions showed significant inhibition of PI3K pathway signaling in glioblastoma tumor tissue following XL765 dosing (36). XL765 is currently in phase I and II clinical studies as a single agent or in combination with other targeted or cytotoxic agents in patients with solid tumors, lymphoma, and leukemia (NCT01390818, NCT01410513, and NCT01403636).

J. Young has ownership interest (including patents) in Exelixis, Inc. No potential conflicts of interest were disclosed by the other authors.

Conception and design: K.-A. Won, K. Yamaguchi, F. Qian, W. Zhang, C.A. Buhr, P. Shen, F.M. Yakes, P. Lamb, P. Foster

Development of methodology: X. Du, J. Wu, K.-A. Won, K. Yamaguchi, P.P. Hsu, F. Qian, C.T. Jaeger, W. Zhang, C.A. Buhr, P. Shen, W. Abulafia, J. Young, F. Chu, M. Lee, S.T. Lam

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): P. Yu, A.D. Laird, X. Du, J. Wu, K.-A. Won, K. Yamaguchi, F. Qian, C.T. Jaeger, W. Abulafia, J. Chen, J. Young, A. Plonowski, F.M. Yakes, S.T. Lam

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P. Yu, A.D. Laird, X. Du, K.-A. Won, K. Yamaguchi, P.P. Hsu, F. Qian, C.T. Jaeger, W. Zhang, W. Abulafia, J. Young, A. Plonowski, M. Lee, F. Bentzien, S.T. Lam, S. Dale, P. Lamb

Writing, review, and/or revision of the manuscript: P. Yu, A.D. Laird, K.-A. Won, W. Zhang, J. Young, F. Bentzien, S.T. Lam, D.J. Matthews, P. Lamb, P. Foster

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): A.D. Laird, K.-A. Won, W. Zhang, W. Abulafia, J. Young, D.J. Matthews

Study supervision: P. Yu, A.D. Laird, K.-A. Won, P.P. Hsu, J. Young, A. Plonowski, F.M. Yakes, D.J. Matthews, P. Lamb, P. Foster

The authors thank Coumaran Egile for critical reading of the article. Requests relating to provision of XL765 (SAR245409) should be directed to Coumaran Egile at sanofi ([email protected]).

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.