Efficacy of the combination of MEK and CDK4/6 inhibitors in vitro and in vivo in KRAS mutant colorectal cancer models

Abstract

Purpose: Though the efficacy of MEK inhibitors is being investigated in KRAS-mutant colorectal cancers (CRC), early clinical trials of MEK inhibitor monotherapy did not reveal significant antitumor activity. Resistance to MEK inhibitor monotherapy developed through a variety of mechanisms converging in ERK reactivation. Since ERK increases cyclin D expression and increases entry into the cell cycle, we hypothesized that the combination of MEK inhibitors and CDK4/6 inhibitors would have synergistic antitumor activity and cause tumor regression in vivo.

Results: The combination of MEK and CDK4/6 inhibitors synergistically inhibited cancer cell growth in vitro and caused tumor regression in vivo in cell line and patient-derived xenograft models. Combination therapy markedly decreased levels of phosphorylated ribosomal protein S6 both in vitro and in vivo and decreased Ki67 staining in vivo.

Experimental design: We performed in vitro proliferation, colony formation, apoptosis, and senescence assays, and Western blots, on a panel of 11 KRAS mutant CRC cell lines treated with the MEK inhibitor MEK162, the CDK4/6 inhibitor palbociclib, or the combination. We also treated 4 KRAS mutant CRC cell line and patient-derived xenografts with the MEK inhibitor trametinib, the CDK4/6 inhibitor palbociclib, or the combination, and performed immunohistochemical and reverse phase protein array analysis.

Conclusions: Combined inhibition of both MEK and CDK4/6 is effective in preclinical models of KRAS mutant CRC and justifies a planned phase II clinical trial in patients with refractory KRAS-mutant CRC.Efficacy of the combination of MEK and CDK4/6 inhibitors in vitro and in vivo in KRAS mutant colorectal cancer models.

Keywords: CDK4/6 inhibitor; KRAS; MEK inhibitor; NRAS.

Figure 1. Combination of MEK and CDK4/6 inhibitors markedly attenuates cell growth in vitro in a panel of KRAS mutant CRC cell lines

(A–F), Colony assays for six representative cell lines treated with the CDK4/6 inhibitor palbociclib (PD), the MEK inhibitor MEK162 (MEK), the combination (MEK/PD), or with DMSO control for 10–14 days. Results are normalized to MEK162 monotherapy. The data represent mean values ± SD for 3–4 independent experiments. (G), Depiction of the stained cell wells. (H), Summary colony assay results for a panel of 11 KRAS mutant CRC cell lines treated with DMSO control, PD, MEK, or MEK/PD at concentrations approximating the IC₅₀ for MEK monotherapy for 10–14 days. Results here are normalized to DMSO control. A waterfall plot is depicted showing values, with cell lines consistently depicted in the same order of T84, LS174T, SW1116, SW948, LS1034, HCT116, SW837, SW403, SW480, Lovo, and SKCO1. The data represent mean values for 3–4 independent experiments. *p < 0.05, *p < 0.006. All p-values were generated by Student's t-test.

Figure 2. Combination of MEK and CDK4/6 inhibitors synergistically attenuates cell growth in a panel of KRAS mutant CRC cell lines

(A), HCT116 cells were treated with the CDK4/6 inhibitor palbociclib (PD) (400 nM, 200 nM, 100 nM, or 50 nM) and/or the MEK inhibitor MEK162 (MEK) (200 nM, 100 nM, 50 nM, or 25 nM respectively) or control for 72 hours and proliferation and cell growth was assessed using MTS assay. The data represent mean values ± SD for 3 independent experiments. (B), Lovo cells treated as described for HCT116. (C), SW480 cells treated as described for HCT116. (D), LS174T cells treated as described for HCT116. (E), Isobologram shows synergy of the combination of PD0332991 and MEK162. Plotted points farther to the bottom left of the Figure represent increasing degree of synergy. *p < 0.05, ♦p < 0.006. All p-values were generated by Student's t-test.

Figure 3. Combination of MEK and CDK4/6 inhibitors inhibits cell growth in KRAS mutant CRC cell lines through a variety of mechanisms

(A), HCT116 cells were treated with the CDK4/6 inhibitor palbociclib (PD, 400 nM), the MEK inhibitor MEK162 (MEK, 200 nM), the combination (MEK/PD), or with DMSO control for 72 hours, and stained with propidium iodide and annexin V-FITC prior to performing flow cytometry. The data represent mean values ± SD for 3 independent experiments. (B), Lovo cells were treated as described for HCT116. C, SW480 cells were treated as described for HCT116. *p < 0.05. ♦p < 0.01 for (A–C), with all p-values generated by Student's t-test. (D), Lovo cells were treated with PD (400 nM), MEK (200 nM), the combination, or with DMSO control for 72 hours, and stained with X-gal to detect senescence-associated β-galactosidase. The proportion of stained cells in 10 random high-powered fields was counted. The data shows mean values from a representative experiment ± SD for the 10 fields. (E), SW480 cells were treated as described for Lovo. ♦p ≤ 0.0002 for combination compared to any other value, with all p-values generated by Student's t-test.

Figure 4. Combination of CDK4/6 and MEK inhibitors induces greater inhibition of phosphorylation of S6 and other growth factor signaling and cell cycle proteins

(A), Protein lysates were obtained from a panel of 9 KRAS mutant CRC cell lines after treatment with DMSO control, palbociclib (PD) 400 nM, MEK162 (MEK) 200 nM, or MEK162 200 nM + palbociclib 400 nM (MEK/PD) for 24 hours, and RPPA was performed. Protein levels were normalized to DMSO controls of each cell line, and a log2 heatmap of the most differentially expressed proteins was generated. (B), Median expression levels of proteins in KRAS mutant CRC cell lines treated with MEK162 (MEK) alone, palbociclib (PD) alone, or the combination (MEK/PD), relative to DMSO-treated control. p-values were determined using paired Student's t-test. (C), Protein expression data from RPPA was mapped according to relevant pathways. Circles represent phosphorylated epitopes on the proteins and squares represent total proteins. Activating KRAS mutations are thought to lead to constitutive activation of downstream signaling pathways, but adding MEK inhibitor and CDK4/6 inhibitor yielded the depicted changes in protein expression. (D), Western blotting of the indicated antibodies for KRAS mutant cell lines HCT116, SW480, SKCO1, Lovo, and SW403 treated with palbociclib 400 nM and/or MEK162 200 nM for 24 hours.

Figure 5. Combination of CDK4/6 and MEK inhibitors induces tumor regression in vivo in KRAS mutant CRC xenografts

(A), Lovo cell line xenografts were treated with vehicle control daily, palbociclib 100 mg/kg daily, trametinib 3 mg/kg every 2 days, or palbociclib 100 mg/kg daily + trametinib 3 mg/kg every 2 days, and tumor volume was measured twice per week. Data are mean ± SEM, with 10 mice/arm. *p < 0.01 on day 21 for combination vs. each other arm by Student's t-test. (B), HCT116 cell line xenografts were treated with control chow or chow containing trametinib + palbociclib, and tumor volume was measured longitudinally. ♦p < 0.001 on day 7 by Student's t-test. (C), SW480 cell line xenografts were treated with control chow or chow containing trametinib + palbociclib, and tumor volume was measured longitudinally. ♦p < 0.001 on day 20 by Student's t-test. (D), Xenografts implanted with patient-derived CRC cells harboring KRAS A146T mutation were treated with vehicle control daily, palbociclib 100 mg/kg daily, trametinib 3 mg/kg every 2 days, or palbociclib 100 mg/kg daily + trametinib 3 mg/kg every 2 days, and tumor volume was measured twice per week. Data are mean ± SEM, with 9–10 mice/arm. ♦p < 0.001 on day 21 for combination vs. any other arm by Student's t-test. Dotted lines represent 50% decrease in tumor volume. (E), Change in tumor volume for each of the patient-derived xenograft mice in D at day 21 of treatment.

Figure 6. Combination of CDK4/6 and MEK inhibitors induces greater inhibition of phosphorylation of S6 and other growth factor signaling and cell cycle proteins in vivo in PDX models of KRAS mutant CRC

(A), Protein lysates were obtained from tumors after 21 days of treatment with vehicle control (C), trametinib (MEK), palbociclib (PD), or trametinib + palbociclib (MEK/PD), and RPPA was performed. Protein levels were normalized to the geometric mean of vehicle-treated controls, and a heatmap of the most differentially expressed proteins was generated. Values are log2-transformed fold-change compared to the geometric mean of the controls. (B), Median expression levels of selected proteins in KRAS mutant CRC PDXs treated with trametinib (MEK) alone, palbociclib (PD) alone, or the combination (MEK/PD) for 21 days, relative to vehicle-treated control. p-values were determined using paired Student's t-test. (C), Percentage of nuclear area stained by Ki67 antibody. Data represent mean ± SD of samples from 3–4 mice. *p ≤ 0.02, ♦p < 0.003 by Student's t-test. (D), Representative immunohistochemistry images for Ki67 staining.