ICEC0942, an Orally Bioavailable Selective Inhibitor of CDK7 for Cancer Treatment

Abstract

Recent reports indicate that some cancer types are especially sensitive to transcription inhibition, suggesting that targeting the transcriptional machinery provides new approaches to cancer treatment. Cyclin-dependent kinase (CDK)7 is necessary for transcription, and acts by phosphorylating the C-terminal domain (CTD) of RNA polymerase II (PolII) to enable transcription initiation. CDK7 additionally regulates the activities of a number of transcription factors, including estrogen receptor (ER)-α. Here we describe a new, orally bioavailable CDK7 inhibitor, ICEC0942. It selectively inhibits CDK7, with an IC₅₀ of 40 nmol/L; IC₅₀ values for CDK1, CDK2, CDK5, and CDK9 were 45-, 15-, 230-, and 30-fold higher. In vitro studies show that a wide range of cancer types are sensitive to CDK7 inhibition with GI₅₀ values ranging between 0.2 and 0.3 μmol/L. In xenografts of both breast and colorectal cancers, the drug has substantial antitumor effects. In addition, combination therapy with tamoxifen showed complete growth arrest of ER-positive tumor xenografts. Our findings reveal that CDK7 inhibition provides a new approach, especially for ER-positive breast cancer and identify ICEC0942 as a prototype drug with potential utility as a single agent or in combination with hormone therapies for breast cancer. ICEC0942 may also be effective in other cancers that display characteristics of transcription factor addiction, such as acute leukaemia and small-cell lung cancer. Mol Cancer Ther; 17(6); 1156-66. ©2018 AACR.

Figure 1. ICEC0942 is a CDK7 selective inhibitor of cancer cell growth.

A, Compound structure of ICEC0942. B, In vitro kinase assays. Inhibition of kinase activity is shown relative to the vehicle treatment, as the mean of 3 experiments; errors bars show SEM. C, Cell lines were treated with increasing concentrations of ICEC0942 for 48 hours. GI₅₀ values are shown for three independent experiments. D, Box and whisker plot (5-95 percentile) showing that ICEC0942 inhibits proliferation of the NCI panel of 60 cancer cell lines. The black dots show cancer cell lines for which GI₅₀ values were outside the 5-95 percentile.

Figure 2. ICEC0942 inhibits phosphorylation of CDK7 substrates, to promote cell cycle arrest and apoptosis.

A, HCT116 cells were treated with ICEC0942 at concentrations shown. Cell lysates were prepared at the indicated time points following ICEC0942 addition. B, Percentage of apoptotic HCT116 cells 24 hours following addition of ICEC0942, determined by Annexin V and propidium iodide staining (n=3 independent experiments; * = p<0.05 relative to vehicle (0) control); t-test. C-D, Cell lysates, prepared 24 hours following addition of ICEC0942 were immunoblotted for PARP cleavage or assayed for caspase 3/7 activity (n=3). E, Flow cytometric analysis was carired out for HCT116 cells 24 hours following addition of ICEC0942 at concentrations shown. Mean percentage of cells in G1, S and G2/M phases are shown for three independent experiments; error bars depict SEM, * = statistically significant (p<0.05) difference between percentage of cells in G1 compared with the vehicle control, # = statistically significant (p<.05) difference between percentage of cells in G2/M, compared with the vehicle control. F, MCF7 cells were arrested in G0/G1, S-phase, or in G2/M. Cells were released from the block by washing and replenishment with fresh medium supplemented with ICEC0942. Shown are the FACS profiles 24 hours following addition of ICEC0942. The numbers below each graph are the percentage of cells in G1, S-phase or G2/M for each enrichment condition for one experiment. These data are included in the time course in Supplementary Figure S3A.

Figure 3. Inhibition of MCF7 tumour xenografts by ICEC0942 is accompanied by reduction in phosphorylation of CDK7 substrates.

A, Mean tumor volumes ±SEM, for randomized nude mice bearing MCF7 tumors treated with 100 mg/kg/day ICEC0942 (n=12 in each arm). Linear regression showed statistically significant (p<0.0001) difference in growth between vehicle and ICEC0942 treated animals. B, FACS of immunostained PBMCs, collected 6 hours after ICEC0942 administration (n=3; *=p<0.05; t-test). C, Immunostaining for PolII, or phosphorylated PolII (1,000 cells/tumor) from vehicle or ICEC0942 treated animals. Representative IHC images are shown. D, Immunoblotting of protein lysates prepared from 3 tumors.

Figure 4. ICEC0942 inhibits growth of colon cancer tumor xenografts.

A-B, Nude mice bearing HCT116 tumors were randomized for PO treatment with 100 mg/kg/daily ICEC0942. Mean tumour volumes ±SEM and mouse weights are shown (n=15). Linear regression analysis showed statistically significant (p<0.0001) difference in growth between vehicle and ICEC0942 treated animals. C, Immunostaining of resected tumors for PolII and PollII Ser2 and Ser5 phosphorylation. Graphed are the scoring of 1000 cells for 3 tumors in each group. D, Immunostaining of PBMCs collected at the end of the study (n=3; *=p<0.05; t-test). E, Biochemistry of bloods collected 6 hours after first or last administration (n=3). F, Blood cell analysis at day 13 of ICEC0942 or vehicle treatment. Also shown are blood counts for animals bearing HCT116 tumors ranging in size from 100-200 mm³ (n=6) and for 3 nude mice without tumors or treatment.

Figure 5. Co-operativity between the CDK7 inhibitor ICEC0942 and hormone therapies in vitro and in vivo.

A, MCF7 cells were treated with ICEC0942 in the presence or absence of the anti-estrogens Tamoxifen or Faslodex over a 12 day period. Growth is shown relative to the vehicle control (n=3). Asterisks represent significant difference (p<0.05) in growth relative to cells grown in the absence of ICEC0942. The hash symbol (#) shows significant (p<0.05) difference in growth between cells cultured in the presence of anti-estrogen and 0.1 µM ICEC0942, compared with cells cultured only in the presence of 0.1 µM ICEC0942. B, Immunoblotting was performed for MCF7 cells treated for 24 hours, with ICEC0942 and/or tamoxifen at the concentrations shown. C, Flow cytometric analysis was carired out for MCF7 cells 24 hours following addition of ICEC0942 and/or tamoxifen, at concentrations shown. Mean percentage of cells in G1, S and G2/M phases are shown for three independent experiments; error bars depict SEM, * = statistically significant (p<0.05) difference between percentage of cells in G2/M compared with the vehicle control, # = statistically significant (p<0.05) difference between percentage of cells in G1, compared with the vehicle control. D, Percentage of apoptotic MCF7 cells 24 hours following addition of ICEC0942 and/or tamoxifen, determined by Annexin V and propidium iodide staining (n=3 independent experiments; * = p<0.05 relative to vehicle (0) control); t-test. E, Animals with MCF7 tumor xenografts treated once daily with vehicle, 100 µg tamoxifen and/or 50 mg/kg ICEC0942 (n=8 for each arm of the study). Multiple comparison test using one-way ANOVA analysis of slopes of linear regression lines was statistically significant (p=0.0002) for the different treatment groups. Shown are the adjusted p-values for treatment pairs from the multiple comparison testing. F, IHC was performed for tumors from 3 animals and scoring done as for Fig. 3. Asterisks show significant differences (p<0.05) from the vehicle treated tumors. All statistical analyses were undertaken using the t-test.