The Preclinical Pharmacology and Therapeutic Activity of the Novel CHK1 Inhibitor SAR-020106

Published OnlineFirst January 12, 2010; DOI: 10.1158/1535-7163.MCT-09-0938

Research Article Molecular Cancer Therapeutics The Preclinical Pharmacology and Therapeutic Activity of the Novel CHK1 Inhibitor SAR-020106

Michael I. Walton1, Paul D. Eve1, Angela Hayes1, Melanie Valenti1, Alexis De Haven Brandon1, Gary Box1, Kathy J. Boxall1, G. Wynne Aherne1, Suzanne A. Eccles1, Florence I. Raynaud1, David H. Williams2, John C. Reader2, Ian Collins1, and Michelle D. Garrett1

Abstract Genotoxic antitumor agents continue to be the mainstay of current cancer chemotherapy. These drugs cause DNA damage and activate numerous cell cycle checkpoints facilitating DNA repair and the main- tenance of genomic integrity. Most human tumors lack functional p53 and consequently have compro-

mised G1-S checkpoint control. This has led to the hypothesis that S and G2-M checkpoint abrogation may selectively enhance genotoxic cell killing in a p53-deficient background, as normal cells would be – rescued at the G1-S checkpoint. CHK1 is a serine/threonine kinase associated with DNA damage linked

S and G2-M checkpoint control. SAR-020106 is an ATP-competitive, potent, and selective CHK1 inhibitor

with an IC50 of 13.3 nmol/L on the isolated human enzyme. This compound abrogates an etoposide-in-

duced G2 arrest with an IC50 of 55 nmol/L in HT29 cells, and significantly enhances the cell killing of gemcitabine and SN38 by 3.0- to 29-fold in several colon tumor lines in vitro and in a p53-dependent fashion. Biomarker studies have shown that SAR-020106 inhibits cytotoxic drug–induced autophosphorylation of CHK1 at S296 and blocks the phosphorylation of CDK1 at Y15 in a dose-dependent fashion both in vitro and in vivo. Cytotoxic drug combinations were associated with increased γH2AX and poly ADP ribose polymerase cleavage consistent with the SAR-020106–enhanced DNA damage and tumor cell death. Irinotecan and gemcitabine antitumor activity was enhanced by SAR-020106 in vivo with minimal toxicity. SAR-020106 represents a novel class of CHK1 inhibitors that can enhance antitumor activity with selected anticancer drugs in vivo and may therefore have clinical utility. Mol Cancer Ther; 9(1); 89–100. ©2010 AACR.

Introduction induces a subsequent delay in the cell cycle to facilitate DNA repair and/or apoptosis (2, 3). The observation that Normal cell division is a highly regulated process in- human cancers frequently have defects in p53 function, volving multiple cyclin-dependent kinases (CDK) and through either direct mutations or the disruption of its their appropriate cyclins that control a series of cell cycle regulatory pathway, with a consequent loss of G1-S checkpoints that ensure the fidelity of DNA replication checkpoint control (3, 4) has led to the concept that G2 (1). Genotoxic antitumor drugs cause DNA damage checkpoint inhibitors combined with DNA-damaging and activate cell cycle checkpoints at G1-S,S,andG2- agents may cause aberrant and lethal mitosis in such tu- M, which are required for DNA repair and the mainte- mors, whereas normal tissue would be rescued at the – nance of genomic integrity (2). The transcription factor p53-dependent G1-S checkpoint (2, 5 7). This strategy p53 is a critical checkpoint protein. In the presence of should therefore increase the therapeutic selectivity of damaged DNA, p53 is stabilized and activated, causing several DNA-damaging anticancer drugs, which are still the upregulation of the CDK inhibitor p21, which the mainstay of current cancer treatment. In normal cells, DNA damage is sensed by a com- plex series of signal transduction pathways. The main Authors' Affiliations: 1Cancer Research UK Centre for Cancer Therapeutics, The Institute of Cancer Research, Haddow Laboratories, proximal kinases are the phosphoinositide 3-kinase Sutton, Surrey, United Kingdom and 2Sareum Ltd, Cambridge, United homologues ATM and ATR (2, 8). These transduce sig- Kingdom nals to the effector kinases CHK1 and CHK2, which Note: Supplementary material for this article is available at Molecular then activate a series of cell cycle checkpoints. CHK1 Cancer Therapeutics Online (http://mct.aacrjournals.org/). is predominantly activated through phosphorylation Corresponding Author: Mike Walton, Cancer Research UK Centre for on S317 and S345 catalyzed by ATR (9, 10), and under- Cancer Therapeutics, The Institute of Cancer Research, Haddow Laboratories, 15 Cotswold Road, Sutton, Surrey, SM2 5NG, United goes autophosphorylation on S296 (11, 12). CHK1 is a Kingdom. Phone: 44-0-20-8722-4322; Fax: 44-0-20-8722-4324. E-mail: serine/threonine kinase that is involved in the S-phase [email protected] checkpoint through stabilizing replication forks, and in doi: 10.1158/1535-7163.MCT-09-0938 the G2 checkpoint control through regulating the stabil- ©2010 American Association for Cancer Research. ity and availability of the CDC25 phosphatases (13–15).

www.aacrjournals.org 89

Walton et al.

Numerous studies have shown that UCN-01, a relatively nonselective inhibitor of CHK1, will preferentially sensitize p53-defective tumor cells to a variety of DNA- damaging anticancer drugs both in vitro and in vivo (16–18). More recently, inhibition of CHK1 function by knockdown of protein expression by small interfering RNA and several more selective CHK1 inhibitors have confirmed these early observations (19–21). As a part of the development of molecularly targeted kinase inhibitors, particularly those that might not have antitumor activity in their own right, it is important to identify appropriate biomarker readouts (22, 23). Consequently, the development of CHK1 inhibitors and the characteriza- Figure 1. Structure of SAR-020106. tion of suitable biomarkers have become an important drug development objective (24–28). In this study, we present the preclinical pharmacology stocks indicated. A2780 cells were stably transfected with and therapeutic utility of SAR-020106, a novel, selective, HPV16E6, and appropriate clones were screened and se- and potent inhibitor of CHK1. We show that SAR- lected essentially as described previously (30). Cells were 020106 can potentiate SN38 and gemcitabine cytotoxi- regularly screened for Mycoplasma using a PCR-based as- in vitro city in several colon cancer cell lines and that say (VenorGem, Minerva Biolabs). this potentiation occurs preferentially in cells that lack p53 function. We further show that SAR-020106 can ab- Kinase Assays rogate an SN38-induced S-phase block and an etopo- CHK1 and CHK2 assay conditions were as described in vitro side-induced S and G2-M arrest .Biomarker previously (31). A kinase profile against 50 different hu- studies showed that SAR-020106 selectively inhibited man kinases was carried out using 10 μmol/L SAR- SN38 or gemcitabine induced autophosphorylation of K 020106 with an ATP concentration equivalent to the m CHK1 at S296, and that this was associated with a loss for each enzyme (Millipore). of CDK1 phosphorylation at Y15, a downstream target γ of CHK1. Increases in H2AX and poly ADP ribose po- Cytotoxicity Assays lymerase (PARP) cleavage were also observed, consis- Cells were plated in 96-well plates (3–6×103 per well) tent with enhanced DNA damage and cell death. or tissue culture flasks (at 2–5×105 per flask) and al- Finally, we present evidence of enhancement of irinote- lowed to attach for 36 h to ensure exponential growth can and gemcitabine therapeutic activity by SAR-020106 at the time of treatment. Cytotoxicity was determined us- in association with CHK1 inhibition in human tumor ing a 96-h (four cell doublings) SRB assay, and GI50 va- xenografts. lues were derived as described previously (32).

Materials and Methods G2 Checkpoint Abrogation Assay Measurements of intracellular CHK1 inhibition were Compounds carried out using HT29 and SW620 in a cell-based ELISA SAR-020106 was synthesized as described previously assay. Cells in 96-well plates (1–2×104 per well) were (29). Clinical formulations of gemcitabine and irinotecan treated with etoposide (50 μmol/L × 1 h) to induce a late were obtained from Eli Lilly (Gemzar, Eli Lilly and S-G2 arrest and then were exposed to different concentra- Pfizer (Campto, Pfizer), respectively. SN38, the active tions of the test compound for a further 21 h in medium metabolite of irinotecan, was purchased from LKT containing nocodazole (100 ng/mL) to trap cells in mito- Laboratories. Other reagents were obtained from sis. Cells were then fixed in 4% formaldehyde in PBS for Sigma-Aldrich Chemical Co. 30 min at 4°C and permeabilized with ice-cold methanol for a further 10 min. Cells were washed in PBS, blocked Cells with 5% milk in PBS × 30 min at 37°C, and then treated HT29, SW620, and Colo205 human colon carcinoma overnight with an antibody to mitotic proteins (MPM-2, cell lines were obtained from American Type Culture Upstate) in 5% milk in TBS. Cells were then rinsed in wa- Collection (American Type Culture Collection lot no. ter containing 0.01% Tween 20 and were treated for 1 h 4487729, 3924081, and 57723824, respectively) and with an Eu-labeled secondary antibody (Eu-N1, Perkin- A2780 ovarian carcinoma cells were obtained from Euro- Elmer) at 37°C. The MPM-2 epitope–expressing (i.e., mi- pean Collection of Animal Cell Cultures (Sigma-Aldrich). totic) cells were quantified using a Wallac Victor 1420 Cells were grown in DMEM containing 10% fetal bovine multilabel counter (Perkin-Elmer). IC50 values for mitosis serum (PAA “Gold”) and 5 mmol/L glutamine in a hu- induction were determined using nocodazole as a posi- midified atmosphere of 5% CO2 at 37°C and passaged tive control. The activity index was used as a measure ≤6 mo before the renewal from frozen, early passage of a compound's ability to induce mitosis relative to its

90 Mol Cancer Ther; 9(1) January 2010 Molecular Cancer Therapeutics

Preclinical Pharmacology of CHK1 Inhibitor SAR-020106

cytotoxicity (i.e., ratio of the G2 checkpoint abrogation alone and GI50 for the CHK1 inhibitor in combination IC50 and SRB cytotoxicity GI50). with a cytotoxic agent.

Potentiation Assay Immunoblotting Cells were exposed to a fixed concentration of the cy- Drug-treated cells were lysed in a buffer containing totoxic agent that inhibited growth by 50% relative to 50 mmol/L HEPES (pH 7.4) 250 mmol/L NaCl, 0.1% untreated controls (GI50 dose) in combination with NP40, 1 mmol/L DTT, 1 mmol/L EDTA, 1 mmol/L increasing concentrations of the CHK1 inhibitor in a 96- NaF, 10 mmol/L β-glycerophosphate, 100 mmol/L h SRB assay. The ability of SAR-020106 to enhance geno- NaVO4, and 1× Complete proteinase inhibitor tablet toxic cell killing was expressed as a potentiation index per 10-mL buffer (Roche). Tumor samples were homoge- (PI), which was the ratio of GI50 for the CHK1 inhibitor nized using a buffer containing 50 mmol/L Tris (pH 7.4),

Figure 2. Effects of SAR-020106 on drug-induced cell cycle arrest in SW620 and HT29 colon cancer cells. A, effects of different concentrations of SAR-020106 alone (0.1, 0.5, and 1 μmol/L) or in combination with a fixed concentration of SN38 (20 nmol/L) for 24 h on cell cycle distribution in SW620 cells. Histograms, cell cycle distribution as measured by propidium iodide (PI) DNA staining; dot plots, DNA synthesis as measured by BrdUrd incorporation.

Cell cycle distribution was assessed by BrdUrd distribution (G1, S, S', and G2-M) as shown and described in Materials and Methods. B, quantification of the effects of different concentrations of SAR-020106 on SN38-induced (20 nmol/L) cell cycle arrest in SW620 cells at 24 h.▪,G1-G0 phase; , S phase; □, S' phase; ,G2-M phase. C, quantification of the effects of different doses of SAR-020106 on etoposide induced (50 μmol/L ×1 h) cell cycle arrest in HT29 cells 23 h following cytotoxic exposure. Symbols as in B. Columns, mean derived from three independent experiments; bars, SD.

www.aacrjournals.org Mol Cancer Ther; 9(1) January 2010 91

Walton et al.

Figure 3. SAR-020106 abrogates a G2 arrest and potentiates the cytotoxicity of gemcitabine and SN38 in vitro. A, G2 checkpoint abrogation assay. HT29 cells in 96-well plates were treated with etoposide (50 μmol/L ×1 h) followed by either no addition▿ ( ), nocodazole▾ ( ; 100 ng/mL), or SAR-020106 (○) at the concentrations indicated, in the presence of nocodazole (100 ng/mL) for 21 h. Cells were also treated with nocodazole alone (100 ng/mL) as a positive control for mitosis (♦). Cells were fixed and stained for the mitotic marker MPM2 as described in Materials and Methods. B, cytotoxicity and potentiation assay. HT29 cells were treated in 96-well plates with either: SAR-020106 alone (○) at the concentrations indicated or, a fixed concentration of SN38▾ ( ;GI50) together with different concentrations of SAR-020106, or a fixed concentration of gemcitabine (•;GI50) combined with different concentrations of SAR-020106. Cells were fixed and stained with SRB after 96 h of drug exposure, and GI50 values determined as described in Materials and Methods. Points, mean for four replicates; bars, SD. Similar results were obtained in repeat experiments.

1 mmol/L NaCl, 1 mmol/L EDTA, 1% Triton X-100, Mice and Antitumor Studies μ 1 mmol/L NaF, 1 mmol/L NaVO4,5 mol/L Fenvalerate, BALB/c mice were obtained from Charles River. Nude nu 5 μmol/L Vbphen, 10 mg/mL TLCK, 1× Complete inhib- mice (CrTac:Ncr-Fox1 ) were bred in-house and mice itor tablet per 10-mL buffer (Roche), protease inhibitor ages 6 to 10 wk were used. Mice were allowed access to cocktail, and phosphatase inhibitor 1 and 2 (Sigma- food (rodent diet no. 5002, Lab Diet) and water ad libitum. Aldrich; ref. 33). Protein concentrations were measured Tumor cells were inoculated s.c. in to both flanks of the by Bradford assay. Proteins samples (50 μg) were dena- recipient mice and palpable tumors were treated at an av- tured in Laemmli buffer and were separated on precast erage diameter of 0.38 to 0.47 cm. Animals were random- 10% or 16% Tris-glycine gels (Novex, Invitrogen). Samples ized into treatment groups of six to eight mice and drugs were transferred to polyvinylidene difluoride membranes, were administered either i.p. or i.v. at 0.01 mL/g as appro- which were incubated in a blocking buffer [5% dried milk priate. SAR-020106 was dissolved in 10% DMSO and 5% in 50 mmol/L Tris (pH 8.0), 150 mmol/L NaCl and 0.1% Tween 20 in saline and was administered i.p. either alone Tween 20] for 1 h, probed with primary antibody in block- or 1 h before or 24 h after the cytotoxic drug administra- ing buffer overnight at 4°C, and detected with HRPO con- tion. Irinotecan was administered on days 0, 7, and 14 at jugated secondary antibody (Bio-Rad). Proteins were 12.5 mg/kg i.p. and SAR-020106 was administered on visualized with enhanced chemiluminescence reagents days 0, 1, 7, 8, 14, and 15 at 40 mg/kg i.p. either alone (Pierce, Thermo Fisher) on Hyperfilm (GE Healthcare) us- or in combination. In the latter case, the CHK1 inhibitor ing a Compact X4 developer (Xograft). Antibodies used was administered 1 h before and 24 h after irinotecan. were as follows: pS296 CHK1, pS317 CHK1, pY15 Similarly, gemcitabine was administered alone on days CDK1, total CDK1, cleaved PARP (CST), Total CHK1 0, 3, 6, and 9 at 60 mg/kg i.v. and SAR-020106 was admin- (SC), pS139 H2AX, total H2AX (Upstate), and glyceralde- istered alone on days 0, 3, 6, and 9 at 40 mg/kg i.p. For hyde-3-phosphate dehydrogenase (Chemicon, Millipore). combination studies, gemcitabine was administered as above except that SAR-020106 was administered either Cell Cycle Analysis 1 h before or 24 h after the cytotoxic agent. Mice were Cells were stained with bromodeoxyuridine (BrdUrd) monitored daily and tumor size and animal weight were and PI as described previously (34). Cells were pulsed measured every 2 or 3 d. Tumor volumes were calculated with 10 μmol/L BrdUrd for 30 min before harvesting from two orthodiagonal measurements and growth and fixation in 70% ethanol. Subsequently, cells were acid delay estimates were derived from growth curves at denatured in 2 mol/L HCl containing trypsin to obtain 300% tumor volume, a value that all tumors grew up to nuclei. These were stained with a primary antibody to or through. Animals were treated in accordance with local BrdUrd (Clone Bu20a, DAKO) and then conjugated to a and national animal welfare guidelines (35). FITC-labeled secondary antibody (Sigma-Aldrich) and stained with PI overnight. Cells were analyzed on a BD Pharmacokinetics LSR II flow cytometer (Becton Dickinson). Cell cycle Drugs were extracted from plasma and tissue homoge- analysis was carried out on gated single-cell events using nates (25% w/v in water) using three volumes of acetoni- WinMDI 2.8. trile containing internal standard. Drug concentrations

92 Mol Cancer Ther; 9(1) January 2010 Molecular Cancer Therapeutics

Preclinical Pharmacology of CHK1 Inhibitor SAR-020106

were determined using reverse-phase high performance 38 other kinases. Inhibition of Flt3, IRAK4, Met, MST2, liquid chromatography and mass spectrometry analysis. p70S6K, Ret, RSK1, and Trk A (<20% activity remaining; Briefly, drugs were separated on a UPLC, BEH, C18 see Supplementary Table 1) was observed. column (Waters), using gradient elution (9 mmol/L ammonium acetate in acetonitrile to 100% acetonitrile SAR-020106 Abrogates Genotoxic-Induced Cell over 3 min and held for 1.6 min) with a flow rate of Cycle Arrest 0.4 mL/min at 55°C. SAR-020106 was identified and A series of experiments were carried out to explore the quantified by mass spectrometry analysis (MRM transi- ability of SAR-020106 to abrogate DNA damage–induced tion of 383.1–86.1 at 15 kV collision energy using an Agi- cell cycle checkpoints. The use of BrdUrd allowed the res- lent 1200 LC system coupled to an Agilent 6410 triple olution of G1 from early S checkpoint and late S from G2- quadrupole mass spectrometer with electrospray ioniza- M checkpoints, which cannot always be resolved using tion in positive ionization mode). Noncompartmental simple PI-based analyses. Initial experiments were car- pharmacokinetic parameters were determined using ried out in SW620 colon carcinoma cells following WinNonLin software version 5.2. SN38 treatment. Figure 2A and B show that there were minimal effects of SAR-020106 alone on cell cycle distri- Statistics bution up to a concentration of at least 1 μmol/L. A con- Statistical significance (*, P < 0.05; **, P < 0.01; and ***, centration of 20 nmol/L SN38 alone caused a profound P < 0.001) was determined using unpaired, one-tailed, S-phase arrest at 24 hours (83.6%) with only 1.8% in G1 t tests as appropriate, with GraphPad Prism 4 software. and 10.7% in G2-M with the remaining 3.9% cells in a noncycling S-phase (S'). The combination of SN38 with SAR-020106 (0.1 μmol/L) caused a marked abrogation Results of this SN38-induced S-phase block (35.8% S phase) with a subsequent increase in G2-M to 59.8%. Increasing con- Potent and Selective Inhibition of CHK1 by centrations of SAR-020106 up to 1 μmol/L seemed to SAR-020106 cause only minor changes in this S-phase population SAR-020106 (Fig. 1) was shown to be a potent inhibitor but a slight decrease in G2-M with a correspondingly in vitro of recombinant human CHK1 with an IC50 of small increase in G1. Consequently, there was clear 13.3 ± 1.3 nmol/L (mean ± SD, n = 3). Ki determination evidence that SAR-020106 caused abrogation of SN38- confirmed that the compound was acting as an ATP mi- induced S-phase arrest. A summary of the cell cycle Ki metic ( = 10.9 nmol/L). The corresponding IC50 values effects obtained in HT29 colon carcinoma cells treated for CHK2 and CDK1 were >10 μmol/L. Kinase profiling with etoposide (50 μmol/L × 1 hour) and subsequently with 10 μmol/L SAR-020106 showed minimal cross-reac- exposed to SAR-020106 for 23 hours is shown in Fig. tivity with Aurora A, CDK1, CDK2, and CHK2 (>75% re- 2C. Etoposide treatment alone induced a late S-phase maining activity compared with 2% for CHK1) and with and G2-M arrest relative to controls (52.8% and 49.3%,

Table 1. Summary of the cytotoxicity and potentiation of genotoxic drugs by the CHK1 inhibitor SAR-020106 in vitro

Cells GI50 (μmol/L) Genotoxic Combination GI50 (μmol/L) PI*

HT29 0.47 ± 0.19 (n = 11) SN38 0.19 ± 0.10 (n = 5) 3.1 ± 1.6† (n =5) Gem 0.18 ± 0.089 (n = 7) 3.0 ± 1.2‡ (n =7) SW620 2.0 ± 0.93 (n = 4) SN38 0.62 ± 0.12 (n = 4) 3.3 ± 0.81§ (n =4) Gem 0.20 ± 0.10 (n = 4) 12 ± 4.9† (n =4) Colo205 3.5 ± 0.57 (n = 3) SN38 1 ± 0.36 (n = 3) 4.1 ± 1.8† (n =3) Gem 0.13 ± 0.036 (n = 3) 29 ± 9.1† (n =3) A2780C 0.45 ± 0.09 (n = 5) SN38 0.37 ± 0.12 (n = 4) 1.4 ± 0.59 (n =4) Gem 0.12 ± 0.0057 (n = 3) 3.6 ± 0.17§ (n =3) A2780E6 0.29 ± 0.034 (n = 5) SN38 0.11 ± 0.054 (n = 4) 3.2 ± 1.4† (n =4) Gem 0.02 ± 0.0081 (n = 3) 16.2 ± 5.6† (n =3)

NOTE: Cytotoxicity (GI50) was determined by SRB assay. See Materials and Methods and Materials for further details.

*PI is the ratio of GI50/combination GI50. Values are mean ± SD of n independent determinations. †P < 0.05 significantly different from unity. ‡P < 0.01 significantly different from unity. §P < 0.001 significantly different from unity.

www.aacrjournals.org Mol Cancer Ther; 9(1) January 2010 93

Walton et al.

Figure 4. Effect of SAR-020106 on SN38 and gemcitabine-induced biomarker changes in SW620 colon cancer cells. A, cells were treated with SN38 (100 nmol/L) alone, with SAR-020106 alone (5 μmol/L), or with a fixed concentration of SN38 (100 nmol/L) in combination with increasing concentrations of SAR-020106 for 24 h. B, cells were treated with gemcitabine (20 nmol/L) alone, with SAR-020106 alone (5 μmol/L), or with a fixed concentration of gemcitabine in combination with different concentrations of SAR-020106 for 24 h. SAR-020106 was administered 1 h before the cytotoxic agent. Protein expression was characterized by Western blotting using 50 μg samples per lane as described in Materials and Methods. CHK1 phosphorylated at S296 (PS296) was used as a biomarker of CHK1 inhibition, pY15 CDK1 as a downstream biomarker of CHK1 inhibition, and pS139 (γ) H2AX and cleaved PARP as biomarkers of DNA damage and apoptosis, respectively. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. Similar results were obtained in repeat experiments.

respectively, in drug-treated cells) and 0.1 μmol/L SAR- a similar degree (Table 1; Fig. 3B), giving a PI of 3. By 020106 caused a profound abrogation of this S-phase block contrast, greater potentiation occurred in SW620 and and a slight increase in the G2-M population (19.1% Colo205 with gemcitabine compared with SN38 (Table 1). and 55.2%, respectively). An increase in concentration to The ability of SAR-020106 to selectively sensitize p53- μ 0.5 mol/L SAR-020106 partially abrogated this G2-M deficient cells to genotoxic chemotherapy was assessed block with a corresponding increase in G1 and S', with min- using an A2780 isogenic cell line in which p53 was imal changes in this distribution occurring at higher doses functionally inactivated by the overexpression of HPV16E6. (1 μmol/L). These data clearly show that SAR-020106 The PI values for parental A2780 (wild-type p53) with SN38 abrogated an etoposide-induced S and G2 arrest. and gemcitabine were 1.4 and 3.6, respectively, whereas the corresponding values in A2780E6 (nonfunctional p53) were −/− G2 Checkpoint Abrogation and Potentiation of 3.2 and 16.2, giving a p53 selectivity of 2.3- and 4.5-fold, Cytotoxicity by SAR-020106 respectively. A cellular assay for measuring functional CHK1 inhibition and G2 checkpoint abrogation was developed. An Inhibition of CHK1 Autophosphorylation and the acute etoposide exposure (50 μmol/L × 1 hour) was used Cell Cycle Pathway by SAR-020106 to arrest cells in a late S-G2 checkpoint before abrogation The characterization of biomarker changes associated with different concentrations of SAR-020106 in the pres- with target inhibition is an important aspect of molecu- ence of nocodazole. This allowed mitotic cells to be col- larly targeted drug development (23). Figure 4A shows lected and quantified in a cell-based ELISA assay. This G2 that phosphorylation at S296 of CHK1, a site of autopho- checkpoint abrogation assay in HT29 and SW620 showed sphorylation, is barely detectable in unperturbed cells, that SAR-020106 was capable of abrogating etoposide-in- but is markedly induced in response to SN38 treatment duced cell cycle arrest with an IC50 of 55 ± 19 nmol/L for 24 hours in SW620 cells. This response is thought to and 91 ± 51 nmol/L, respectively (mean ± SD, n =5; be catalyzed by the DNA damage–induced activation of Fig. 3A). Table 1 shows that SAR-020106 was relatively CHK1 providing a potential biomarker of CHK1 activity μ nontoxic with a GI50 of 0.48 mol/L in HT29 and (11, 12). The combination of SN38 with increasing con- 2 μmol/L in SW620, resulting in an activity index of 8.7 centrations of SAR-020106 was shown to decrease this and 22, respectively. Furthermore, it enhanced the cyto- signal in a concentration-dependent fashion with com- toxicity of both SN38 and gemcitabine in HT29 cells to plete suppression at ≥1 μmol/L SAR-020106. These

94 Mol Cancer Ther; 9(1) January 2010 Molecular Cancer Therapeutics

Preclinical Pharmacology of CHK1 Inhibitor SAR-020106

changes in CHK1 phosphorylation were associated with tent with the loss of the pS296 CHK1 signal. In addition, minimal alterations in total protein expression. As shown the band shift of CDK1, which is indicative of Y15 phos- by the previous data, a critical response of cells to DNA phorylation, was also inhibited at concentrations of SAR- damage is cell cycle arrest. In particular, phosphorylation 020106 that are ≥1 μmol/L. Interestingly, combinations of tyrosine 15 on CDK1 (pY15) blocks the activity of this of SN38 and SAR-020106 at ≥0.1 μmol/L caused an in- γ kinase, causing arrest at the G2-M transition. CHK1 con- crease in S139 H2AX phosphorylation ( H2AX), a signal trols this arrest through regulating CDC25 phosphatase associated with DNA double-strand breaks (38) and activity (14, 36, 37), and consequently, pY15 CDK1 can there was a corresponding increase in PARP cleavage, a be regarded as a downstream biomarker for CHK1 activ- marker of cell death (39), at concentrations ≥1 μmol/L ity and G2-M arrest. Figure 4A shows that SN38 caused SAR-020106. Figure 4B shows that a concentration of an increase in pY15 CDK1 consistent with a cell cycle 20 nmol/L gemcitabine (which induced pS296 CHK1 arrest, whereas SAR-020106 alone had no effect. By con- and pY15 CDK1) combined with SAR-020106 for 24 trast, the combination of SN38 and SAR-020106 blocked hours in SW620 cells caused similar effects to combina- the induction of pY15 at ≥1 μmol/L SAR-020106, consis- tions with SN38, although S296 CHK1 phosphorylation

Figure 5. Pharmacokinetic properties and pharmacodynamic effects of SAR-020106 in combination with irinotecan in vivo. A, pharmacokinetics of SAR-020106 in nontumor-bearing BALB/c mice following 5 mg/kg i.p. bolus administration in plasma (○) and muscle tissue (•). Points, mean for three mice per time point; bars, SD. Drug concentrations and pharmacokinetic parameters were determined as described in Materials and Methods. B, concentrations of SAR-

020106 occurring in plasma (white columns) and SW620 tumors (black columns) 6 and 24 h following initial drug administration. Dashed line, the IC50 concentration of SAR-020106 required for cellular G2 checkpoint abrogation determined in vitro. Tumor-bearing nude mice were given SAR-020106 (40 mg/kg i.p.) 1 h before a fixed dose of irinotecan (50 mg/kg i.p.). Columns, mean for three mice per time point; bars, SD. C, biomarker changes occurring in SW620 tumor xenografts from B 24h following treatment with either irinotecan alone or in combination with SAR-020106. Tumor-bearing animals were given either vehicle alone, irinotecan alone (25 or 50 mg/kg i.p.), or combined with a fixed dose of SAR-020106 (40 mg/kg i.p.) 1 h before irinotecan administration. Animals were sacrificed 6 or 24 h following initial drug treatment and tumors were recovered and snap frozen. Tumor protein expression was characterized by Western blotting using 50 μg samples per lane as described in Materials and Methods. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. Similar results were obtained in repeat experiments. D, densitometric quantification of the immunoblots shown in C.▪, pS296 CHK1; □, total CHK1; , pY15 CDK1; and , total CDK1. Statistics: *, P < 0.05; ***, P < 0.001 significantly different from irinotecan treatment alone.

www.aacrjournals.org Mol Cancer Ther; 9(1) January 2010 95

Walton et al.

Table 2. Summary of SAR-020106 pharmacokinetics in BALB/c mice following 5 mg/kg i.v. or i.p.

Route Tissue Tmax (h) Cmax (nmol/L) T1/2 (h) Cl (L/h) Vz (L) AUC0-∞ (nmol/L h)

I.V. Plasma N/A 1350 0.658 0.371 0.353 703 I.P. Plasma <0.083 415 0.995 0.781 1.12 334 I.P. Muscle 0.5 644 5.59 0.100 0.806 2615

NOTE: Nonparametric pharmacokinetic parameters were determined using PC Nonlin software with six to seven time points and two to three mice per time point. Abbreviation: N/A, not available.

seemed to be slightly more sensitive to SAR-020106 inhi- at least 24 hours, consistent with checkpoint abrogation bition and γH2AX was induced ≥1 μmol/L SAR-020106. in tumor cells in vivo (see Fig. 5B). Plasma binding studies Similar results were obtained in HT29 cells following established that SAR-020106 was 94.4% bound to mouse 24 hours of treatment with SN38 or gemcitabine (data plasma proteins. not shown). These results show that SAR-020106 is an inhibitor of genotoxic drug–induced CHK1 activation in Inhibition of DNA Damage–induced CHK1 human tumor cells in vitro. Biomarkers by SAR-020106 In vivo Having shown biomarker modulation in vitro and Pharmacokinetics of SAR-020106 in Mice established that SAR-020106 exhibited pharmacoki- The preliminary pharmacokinetic properties of SAR- netics that were consistent with potentially active tu- 020106 were determined in nontumor-bearing BALB/c mor drug exposures, SAR-020106 was combined mice to ensure that tissue drug concentrations were ade- with the topoisomerase 1 inhibitor irinotecan to explore quate for CHK1 inhibition in vivo. Figure 5A and Table 2 its effects on CHK1 biomarker expression in vivo. summarize the pharmacokinetic characteristics of SAR- Figure 5C and D show that irinotecan at 25 and 020106 following administration at 5 mg/kg either i.v. 50 mg/kg i.p. caused a marked increase in CHK1 au- or i.p. Peak plasma concentration was 1.35 μmol/L after tophosphorylation at S296 in SW620 tumors, 24 hours i.v. administration with a relatively short half-life and following drug administration, compared with controls. high clearance and volume of distribution giving an In addition, irinotecan also enhanced Y15 CDK1 phos- AUC0-∞ of 703 nmol/L h. Following i.p. administration, phorylation consistent with a drug-induced cell cycle the plasma elimination time course was comparable with arrest in these tumor cells at 24 hours. The addition that following i.v. administration (data not shown), but of SAR-020106 at 40 mg/kg caused a significant reduc- the peak drug concentration was 3-fold lower, the clear- tion in the pS296 CHK1 signal at 24 hours with mini- ance was 2-fold slower, and the volume of distribution mal effects on total CHK1 levels consistent with CHK1 in vivo was 4-fold higher (Table 2). This resulted in an AUC0-∞ inhibition (Fig. 5D). Moreover, there was also a of 334 nmol/L h giving, an i.p. bioavailability of 47.5%. significant reduction in pY15 CDK1 signal intensity Interestingly, following i.p. administration, muscle tissue following SAR-020106 treatment, suggesting that irino- uptake was higher than in plasma and there was evi- tecan cell cycle arrest may have been abrogated in vivo dence of tissue retention as shown by the 5- to 6-fold lon- (Fig. 5D). ger half-life and correspondingly lower clearance. Consequently, muscle drug exposure was much higher Antitumor Effects of SAR-020106 in Combination than in plasma with an AUC0-∞ of 2615 nmol/L h and with Irinotecan or Gemcitabine a tissue/plasma ratio of 8:1. Other tissues showed tis- The ability of SAR-020106 to potentiate the antitumor sue/plasma ratios of 10:1 to 50:1 following i.v. adminis- activity of irinotecan and gemcitabine in vivo was as- tration, consistent with tissue drug accumulation and sessed in SW620 xenografts. Figure 6A shows that retention at this dose. Assuming linear pharmacokinetics, there were minimal effects of either irinotecan or a dose of 40 mg/kg i.p. would be expected to result in SAR-020106 alone in SW620 tumors at the doses used. tissue exposures of 21 to 130 μmol/L h, which equals The mean time for control tumors to reach 300% of or exceeds that required for biomarker modulation their initial treatment size was 6.2 days. By comparison in vitro (24 μmol/L h; see Fig. 3). Figure 5B shows that SAR-020106 alone–treated tumors (40 mg/kg i.p.) and SAR-020106 tumor concentrations greatly exceeded plas- irinotecan alone–treated tumors (12.5 mg/kg i.p.) ma drug concentrations with tumor/plasma ratios of reached this size after a mean of 6.3 and 6.9 days, 47:1 and 85:1 at 6 and 24 hours, respectively, following (P = 0.159 and 0.460, respectively). However, there 40 mg/kg i.p. Moreover, this dose gave tumor drug was a clear decrease in tumor growth associated concentrations that greatly exceeded those required for with the combination with tumors reaching 300% by in vitro G2 checkpoint abrogation (IC50, 55 nmol/L) for 12.5 days, resulting in a significant growth delay relative

96 Mol Cancer Ther; 9(1) January 2010 Molecular Cancer Therapeutics

Preclinical Pharmacology of CHK1 Inhibitor SAR-020106

to irinotecan alone of 5.6 days (P < 0.01) and an in- Discussion crease in growth delay of 5.5 days. Two mice were culled on days 15 and 18 in the SAR-020106 treatment These studies were undertaken to characterize the pre- alonegroupduetotumorgrowth.Therewasminimal clinical pharmacology of SAR-020106 and to identify body weight loss associated with this combination genotoxic agents, which in combination with SAR- giving a nadir on day 11, and a mean body weight 020106 exhibited enhanced antitumor activity in human of 96.4% of the initial weight. The combination of tumor xenografts. As a part of these studies, it was im- gemcitabine with SAR-020106 in SW620 confirmed portant to identify and monitor suitable biomarkers of that SAR-020106 alone at 40 mg/kg i.p. had negligible CHK1 inhibition and to establish any pharmacokinetic- antitumor activity (Fig. 6B), with a mean time for tu- pharmacodynamic relationships both in vitro and in vivo. mors to reach 300% of their initial treatment size of This is a particularly important aspect of CHK1 inhibitor 7.0 compared with 6.2 days in controls (P = 0.204). development as they are unlikely to exhibit any antitu- Gemcitabine alone had a clear antitumor effect with mor activity as single agents. The identification and use tumors taking 10.1 days to reach 300%, giving a sig- of these biomarkers allow the integrity of the CHK1 in- nificant growth delay of 3.9 days relative to controls hibitor/target pathway to be monitored and maintained (P < 0.001). The combination of gemcitabine and throughout the preclinical and clinical evaluation pro- SAR-020106 was more potent when the two agents cess. Furthermore, this approach allows the construction were administered simultaneously rather than with a of a “pharmacologic audit trail” that will facilitate fur- 24-hour delay following the cytotoxic. The mean time ther drug development and interpretation of clinical for tumors to reach 300% of the initial treatment trials (23). size was 14.7 and 12.5 days for the simultaneous SAR-020106 is a novel, selective, and potent CHK1 in- and delayed combination, respectively, resulting in hibitor. This compound was initially assayed against re- growth delays of 8.5 and 6.3 days relative to control combinant human CHK1 and was shown to be a potent and a significant increase in growth delay of 4.6 days ATP competitive inhibitor. Kinase profiling data con- (P < 0.01) and 2.4 days (P = 0.0514) relative to gem- firmed that there was minimal activity against the func- citabine alone. This represents an increase in growth tionally related kinase CHK2 and the cell cycle kinase delay of 3.8 days for the simultaneous treatment, CDK1 (>7,000-fold selectivity). This was considered which is a doubling of the antitumor effect of gemci- advantageous as the therapeutic utility of CHK2 is still tabine alone. One mouse died in the simultaneous unclear (40), and inhibiting CDK1 would interfere with combination group on day 18. Once again this syner- cell cycle progression and prevent G2 checkpoint abroga- gistic antitumor interaction was associated with mini- tion (41). There was some cross-reactivity with several mal weight loss with a nadir on day 16 and a mean receptor tyrosine kinases such as Flt3, Ret, Met, and Trk body weight of 93.3% of the initial weight. A, but it has been argued that sustained inhibition of

Figure 6. Antitumor effects of combining SAR-020106 with different cytotoxic agents in nude mice bearing SW620 xenograft tumors. A, combination of SAR-020106 with irinotecan. Vehicle alone (•) and irinotecan alone (◊; 12.5 mg/kg i.p.) were administered on the days shown by arrows.▿, SAR-020106 alone (40 mg/kg i.p.);▾, irinotecan and SAR-020106 combined (SAR-020106 was administered 1 h before irinotecan using the single agent doses). B, combination of SAR-020106 with gemcitabine. •, vehicle alone; ◊, gemcitabine alone (60 mg/kg i.v.);▿, SAR-020106 alone (40 mg/kg i.p.); ♦, gemcitabine and SAR-020106 combined at single agent doses (SAR-020106 1 h before gemcitabine); or▾, combination with SAR-020106 administered 24 h following gemcitabine. Points, mean (n =10–16); bars, SEM. Tumor size and animal body weight was assessed as described in Materials and Methods.

www.aacrjournals.org Mol Cancer Ther; 9(1) January 2010 97

Walton et al.

receptor tyrosine kinases is required for a pharmacologic checkpoint abrogation for gemcitabine and a CHK1 in- effect and is therefore unlikely to occur with the relatively hibitor (PD-321852) in pancreatic cancer cells (47). Conse- short exposures required for checkpoint abrogation (25). quently, the cellular response to a genotoxic agent and Abrogation of an etoposide-induced G2 checkpoint CHK1 inhibitor combination may depend on numerous and induction of mitosis was quantified in a cell-based factors including the type of genotoxic agent used, the ELISA assay. These data together with the cell cycle re- checkpoint response involved, the DNA repair capacity, sults (Fig. 2) confirmed that SAR-020106 abrogates an and the genetic background of the tumor cell. One fasci- etoposide-induced G2 arrest in HT29 and SW620 cells nating corollary to this is the recent observation that tu- with similar potencies, which were 9- and 22-fold lower mor cyclin B1 expression may predict efficacy for CHK1 than the corresponding GI50 values giving a potential inhibitors (46). therapeutic margin. The apparent discrepancy between The pharmacokinetics of SAR-020106 showed that the the similar cellular CHK1 kinase inhibition and the dif- drug was concentrated and retained in tissues with a ferent cytotoxicity of SAR-020106 in these two cell lines large AUC0-∞ relative to plasma. Consequently, tumor suggests that cytotoxicity is not directly associated with concentrations exceeding those required for CHK1 inhi- CHK1 inhibitory activity. Similar checkpoint abrogation bition were readily achieved following a dose of activity was reported for PF00477736 (25), with the dual 40 mg/kg i.p. The protein binding was moderately high CHK1 and CHK2 kinase inhibitor AZD7762 giving a but did not seem to prevent active drug concentrations slightly lower value (24). The ability of SAR-020106 to from being achieved in vivo, and tissue drug accumula- potentiate the cytotoxicity of SN38 (the active metabo- tion seemed unimpaired. Despite these observations, it lite of irinotecan, a topoisomerase 1 inhibitor) and gem- ispossiblethatactivedrugconcentrationsmaybe citabine (an antimetabolite) was evaluated in several required for at least one cell cycle in tumors and either colon tumor cell lines in vitro. In general, the cytotoxi- repeat drug administration or drug infusion may be city of gemcitabine was potentiated to a greater extent necessary for maximum antitumor effects in vivo. than SN38, although the degree of potentiation may de- Biomarker studies with SN38 and gemcitabine pend on compound concentration and schedule as well in vitro showed enhanced CHK1 autophosphorylation as genetic background. Importantly, the use of an E6- on S296 consistent with DNA damage–induced CHK1 transfected cell line with compromized p53 function, activation (12). SAR-020106 inhibited this effect in a confirmed that CHK1 inhibition by SAR-020106 selec- concentration-dependent manner with complete inhibi- − − tively enhanced genotoxic cell killing in a p53 / depen- tion of >1 μmol/L × 24 hours in both colon tumor cell dent fashion, with gemcitabine once again showing lines. Tyrosine 15 CDK1 phosphorylation was also consistently greater sensitization than SN38 in this modulated in a similar fashion, indicating that CHK1- model. Other studies have also shown that several mediated abolition of genotoxic drug induced cell cycle CHK1 inhibitors can potentiate SN38 and gemcitabine arrest. Biomarker changes consistent with CHK1 inhibi- cytotoxicity in vitro with gemcitabine exhibiting the tion contributing to increased DNA damage and cell greatest potentiation (24–27). death specifically in the SAR-020106 treated combina- At present, it is unclear which cytotoxic agents are tions were also detected. Perhaps more significantly, ir- most efficacious in combination with CHK1 inhibition. inotecan readily induced S296 CHK1 and pY15 CDK1 Nevertheless, the genotoxic agents that are most promi- phosphorylation in tumor xenografts after 24 hours of nently reported in publications on CHK1 inhibition and exposure, confirming CHK1 activation and a cell cycle show marked potentiation are irinotecan and gemcita- arrest. SAR-020106 was able to inhibit these effects, bine. Both of these agents are reported to be involved confirming that CHK1 inhibition and abrogation of in activating the intra–S-phase checkpoint through thecellcyclearrestcouldbeachievedin vivo under stalled replication forks (42, 43), and CHK1 is required these conditions. Subsequent antitumor studies with rel- for stabilizing these forks and preventing chromosomal atively modest doses of irinotecan and gemcitabine breaks as well as reinitiating replication (13, 44). Our cell showed that SAR-020106 could clearly potentiate the cycle studies showed that both SN38 and etoposide antitumor effects of these genotoxic agents in a syner- caused S-phase arrest that was completely abrogated by gistic fashion. These results support the use of pS296 SAR-020106. Moreover, in vitro potentiation experiments CHK1 and 15Y CDK1 as biomarkers of CHK1 inhibi- and biomarker studies clearly show that this effect was tion and confirm the relationship between CHK1 inhibi- associated with increased cell killing and induction of tion and the enhanced antitumor activity of SN38 γH2AX and PARP cleavage with SN38. Although a cor- in vivo. Reassuringly, there was minimal weight loss relation between CHK1 inhibition and premature mitosis or antitumor effects of SAR-020106 alone, despite re- and enhanced cell killing has been reported for camp- ports of the lethal effects of CHK1 knockouts in mouse tothecins and gemcitabine (45, 46), there is still some embryos (48). The combination of either genotoxic controversy about the exact mechanism of increased cy- agent with SAR-020106 was also well tolerated with totoxicity. For example, it has recently been reported that minimal body weight loss, suggesting that intermittent there is a better correlation between inhibition of the inhibition of CHK1 may have minimal toxic side effects. Rad 51 repair protein response and sensitization than Interestingly our studies showed that the simultaneous

98 Mol Cancer Ther; 9(1) January 2010 Molecular Cancer Therapeutics

Preclinical Pharmacology of CHK1 Inhibitor SAR-020106

combination of gemcitabine with SAR-020106 was more Eccles, F.I. Raynaud, I. Collins, and M.D. Garrett are efficacious than gemcitabine followed 24 hours later by employees of The Institute of Cancer Research that has SAR-020106. Although we have shown that CHK1 inhi- a commercial interest in CHK1 inhibitors. J.C. Reader is bition by SAR-020106 gives a synergistic antitumor re- an employee of Sareum Ltd, which has a commercial in- sponse with irinotecan and gemcitabine, an equally terest in CHK1 inhibitors. Sareum PLC is a wholly critical question relates to the ability of CHK1 inhibition owned subsidiary of Sareum Holdings PLC, of which J. to facilitate tumor cures. Other studies have shown cy- C. Reader is a shareholder. D.H. Williams is a former totoxic anticancer drug potentiation with different employee of Sareum Ltd and is a shareholder of Sareum CHK1 inhibitors (24–27) and there is also evidence of Holdings PLC. Both Sareum and the Institute of Cancer increased cure rates (24). The mechanism by which Research have been involved in a commercial collabora- CHK1 inhibitors enhance anticancer drug cell killing tionwithCancerResearchTechnologyLtdtodiscover is still unclear and the role and efficiency of checkpoint and develop inhibitors of CHK1. abrogation versus the inhibition of DNA damage repair remains to be resolved. Acknowledgments In conclusion, this study shows that SAR-020106 is a selective, potent CHK1 inhibitor. The cytotoxicity of SN38 We thank Gowri Vijayaraghavan for the help with the and gemcitabine can be enhanced in several different flow cytometry studies and the Cell Cycle Control Team human tumor cell lines in vitro by SAR-020106 and this for useful discussions. is associated with biomarker changes and cell cycle effects consistent with CHK1 inhibition. Perhaps most impor- Grant Support tantly, we show that SAR-020106 can enhance the antitumor effects of both irinotecan and gemcitabine Cancer Research UK [CUK] grant number C309/ in vivo with appropriate biomarker changes and minimal A8274. We also acknowledge NHS funding to the NIHR toxicity. These results support further CHK1 inhibitor Biomedical Research Centre. development and clinical evaluation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must Disclosure of Potential Conflicts of Interest therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. M.I.Walton,P.D.Eve,A.Hayes,M.Valenti,A.De Received 10/9/09; accepted 11/3/09; published Haven Brandon, G. Box, K.J. Boxall, G.W. Aherne, S.A. OnlineFirst 1/6/10.

References 1. Sherr CJ. Cancer cell cycles. Science 1996;274:1672–7. phase checkpoint monitoring replication origin firing. J Cell Biol 2. Zhou BB, Elledge SJ. The DNA damage response:putting check- 2001;154:913–23.

points in perspective. Nature 2000;408:433–9. 14. Xiao Z, Chen Z, Gunasekera AH, et al. CHK1 mediates S and G2 3. Oren M. Regulation of the p53 tumor suppressor protein. J Biol arrests through CDC25A degradation in response to DNA-damaging Chem 1999;274:36031–4. agents. J Biol Chem 2003;278:21767–73. 4. Hollstein M, Sidransky D, Vogelstein B, Harris CC. P53 mutations in 15. Peng CY, Graves PR, Thoma RS, Wu Z, Shaw AS, Piwinca-Worms

human cancer. Science 1991;253:49–53. H. Mitotic and G2 checkpoint control: regulation of 14-3-3 protein 5. Bucher N, Britten CD. G2 checkpoint abrogation and checkpoint binding by phosphorylation of CDC25C on serine 216. Science kinase-1 targeting in the treatment of cancer. Br J Cancer 2008;98: 1997;277:1501–5. 523–8. 16. Lee SI, Brown MK, Eastman A. Comparison of 7-hydroxystaurospor-

6. Kawabe T. G2 checkpoint abrogators as anticancer drugs. Mol ine (UCN-01) and other staurosporine analogs to abrogate cisplatin- Cancer Ther 2004;3:513–9. induced cell cycle arrest in human breast cancer cell lines. Biochem 7. Okada H, Mak TW. Pathways of apoptotic and non-apoptotic death Pharmacol 1999;58:1713–21. in tumor cells. Nat Rev Cancer 2004;4:592–9. 17. Wang Q, Fan S, Eastman A, Worland PJ, Sausville EA, O'Connor PM.

8. Abraham RT. Cell cycle checkpoint signalling through the ATM and UCN-01: potent abrogator of G2 checkpoint function in cancer cells ATR kinases. Genes Dev 2001;15:2177–96. with disrupted p53. J Natl Cancer Inst 1996;88:956–65. 9. Zhao H, Piwnica-Worms H. ATR-mediated checkpoint pathways 18. Graves PR, Yu L, Schwarz JK, et al. The CHK1 protein kinase and the regulate phosphorylation and activation of human CHK1. Mol Cell CDC25C regulatory pathways are targets of the anticancer agent Biol 2001;21:4129–39. UCN-01. J Biol Chem 2000;275:5600–5. 10. Guo ZJ, Kmagai A, Wang SX, Dunphy WG. Requirement for ATR in 19. Ganzinelli M, Carrassa L, Crippa F, Tavecchio M, Broggini M, phosphorylation of CHK1 and cell cycle regulation in response to Damia G. Checkpoint kinase 1 down-regulation by an inducible DNA replication blocks and UV-damage in Xenopus egg extracts. small interfering RNA expression system sensitized in vivo Genes Dev 2000;14:2745–56. tumors to treatment with 5-fluorouracil. Clin Cancer Res 2008; 11. Smits VAJ. Spreading the signal—dissociation of CHK1 from 14:5131–41. chromatin. Cell Cycle 2006;5:1039–43. 20. Janetka JW, Ashwell S, Zabludoff S, Lyne P. Inhibitors of checkpoint 12. Clarke CA, Clarke PR. DNA-dependent phosphorylation of CHK1 kinases: from discovery to the clinic. Curr Opin Drug Discov Devel and claspin in human cell-free system. Biochem J 2005;388: 2007;10:473–86. 705–12. 21. Tse AN, Carvajal R, Scwartz GK. Targeting checkpoint kinase 1 in 13. Feijoo C, Hall-Jackson C, Wu R, et al. Activation of a mammalian cancer therapeutics. Clin Cancer Res 2007;13:1955–60. CHK1duringDNAreplicationarrest:roleofCHK1intheintra-S 22. Park JW, Kerbel RS, Kelloff GJ, et al. Rationale for biomarkers and

www.aacrjournals.org Mol Cancer Ther; 9(1) January 2010 99

Walton et al.

surrogate end points in mechanism driven oncology drug develop- 35. Workman P, Twentyman P, Balkwill F, Balmain A, Chaplin D, Double ment. Clin Cancer Res 2004;10:3885–96. J. United Kingdom co-ordinating committee on cancer research 23. Workman P. How much gets there and what does it do? The need guidelines for the welfare of animals in experimental neoplasia. 2nd for better pharmacokinetic and pharmacodynamic end points in ed. Br J Cancer 1998;77:1–10. contemporary drug discovery and development. Curr Pharm Des 36. Watanabe N, Broome M, Hunter T. Regulation of the human WEE1- 2003;9:891–902. Hu CDK tyrosine 15-kinase during the cell cycle. EMBO J 1995;14: 24. Zabludoff SD, Grondine MR, Sheehy AM, et al. AZD7762, a nov- 1878–91. el checkpoint kinase inhibitor, drives checkpoint abrogation and 37. Wang Y, Li J, Booher RN, et al. Radiosensitization of p53 mutant

potentiates DNA-targeted therapies. Mol Cancer Ther 2008;7: cells by PD0166285, a novel G2 checkpoint abrogator. Cancer Res 2955–66. 2001;61:8211–7. 25. Blasina A, Hallin J, Chen E, et al. Breaching the DNA damage 38. Rogakou EP, Pilch DR, Orr AH, Ivanova VS, Bonner WM. DNA checkpoint via PF-00477736, a novel small-molecule inhibitor of double-strand breaks induce histone H2AX phosphorylation on checkpoint kinase 1. Mol Cancer Ther 2008;7:2394–404. serine 139. J Biol Chem 1998;273:5858–68. 26. Tse AN, Rendahl KG, Sheikh T, et al. CHIR-124, a novel potent 39. Gobeil S, Boucher CC, Nadeau D, Poirier GG. Characterization of inhibitor of CHK1, potentiates the cytotoxicity of topoisomerase 1 the necrotic cleavage of poly (ADP-ribose) polymerase (PARP-1): poisons in vitro and in vivo. Clin Cancer Res 2007;13:591–602. implication of lysosomal proteases. Cell Death Diff 2001;8:588–94. 27. Matthews DJ, Yakes FM, Chen J, et al. Pharmacological abrogation 40. Antoni L, Sodha N, Collins I, Garrett MD. CHK2 kinase: cancer of S-phase checkpoints enhances the antitumor activity of gemcita- susceptibility and cancer therapy—two sides of the same coin? bine in vivo. Cell Cycle 2007;6:104–10. Nat Rev Cancer 2007;7:925–36. 28. Wang GT, Gaoquan L, Mantei RA, et al. 1(5-Chloro-2-alkoxyphenyl)- 41. Shapiro GI. Cyclin dependent kinase pathways as targets for cancer 3-(5-cyano-pyrazi-2-yl)ureas as potent and selective inhibitors of treatment. J Clin Oncol 2006;24:1770–83. CHK1 kinase: synthesis, preliminary SAR, and biological activities. 42. Pommier Y. Topoisomerase 1 inhibitors: camptothecins and beyond. J Med Chem 2005;48:3118–21. Nat Rev Cancer 2006;6:789–802. 29. Collins I, Reader JC, Williams DH, et al. Int Pat Appl WO 103966, 43. Ewald B, Sampath D, Plunkett W. H2AX phosphorylation marks 2009. Chem Abstr 2009;151:1042183. gemcitabine-induced replication forks and their collapse upon 30. Pestell KE, Hobbs SM, Titley JC, Kelland LR, Walton MI. Effect of S-phase checkpoint abrogation. Mol Cancer Ther 2007;6:1239–48. p53 status on sensitivity to platinum complexes in a human ovarian 44. Petermann E, Caldecott KW. Evidence that the ATR/CHK1 pathway cancer cell line. Mol Pharmacol 2000;57:503–11. maintains normal replication fork progression during unperturbed 31. Matthews TP, Klair S, Burns S, et al. Identification of inhibitors of S phase. Cell Cycle 2006;5:2203–9. checkpoint kinase 1 through template screening. J Med Chem 45. Morgan MA, Parsels LA, Parsels JD, Lawrence TS, Maybaum J. The 2009;52:4810–9. relationship of premature mitosis to cytotoxicity in response to 32. Skehan P, Storeng R, Scudiero D, et al. New colorimetric cytotoxicity checkpoint abrogation and antimetabolite treatment. Cell Cycle assay for anticancer-drug screening. J Natl Cancer Inst 1990;82: 2006;5:1983–8. 1107–12. 46. Xiao Z, Xue J, Gu W-Z, et al. Cyclin B1 is an efficacy-predicting 33. Gowan SM, Hardcastle A, Hallsworth AE, et al. Application of biomarker for CHK1 inhibitors. Biomarkers 2008;13:579–96. meso scale technology for the measurement of phosphoproteins 47. Parsels LA, Morgan MA, Tanska DM, et al. Gemcitabine sensitization in human tumor xenografts. Assays Drug Dev Technol 2007;5: by checkpoint kinase 1 inhibition correlates with inhibition of a Rad51 391–401. DNA damage response in pancreatic cancer cells. Mol Cancer Ther 34. Wilson GD. Analysis of DNA—measurements of cell kinetics by the 2009;8:45–54. bromodeoxyuridine/antibromodeoxyuridine method. In: Ormerod 48. Takai H, Tominaga K, Motoyama N. Aberrant cell cycle checkpoint MG, editor. Flow Cytometry (2nd Edition): A practical approach. function and early embryonic death in CHK1(−/−) mice. Genes Dev Oxford: IRL Press; 1994, p. 137–56. 2000;14:1439–47.

100 Mol Cancer Ther; 9(1) January 2010 Molecular Cancer Therapeutics

The Preclinical Pharmacology and Therapeutic Activity of the Novel CHK1 Inhibitor SAR-020106

Michael I. Walton, Paul D. Eve, Angela Hayes, et al.

Mol Cancer Ther 2010;9:89-100. Published OnlineFirst January 12, 2010.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-09-0938

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2012/06/01/1535-7163.MCT-09-0938.DC1

Cited articles This article cites 47 articles, 24 of which you can access for free at: http://mct.aacrjournals.org/content/9/1/89.full#ref-list-1

Citing articles This article has been cited by 12 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/9/1/89.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/9/1/89. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.