A preclinical evaluation of the MEK inhibitor refametinib in HER2-positive breast cancer cell lines including those with acquired resistance to trastuzumab or lapatinib

Abstract

Purpose: The MEK/MAPK pathway is commonly activated in HER2-positive breast cancer, but little investigation of targeting this pathway has been undertaken. Here we present the results of an in vitro preclinical evaluation of refametinib, an allosteric MEK1/2 inhibitor, in HER2-positive breast cancer cell lines including models of acquired resistance to trastuzumab or lapatinib.

Methods: A panel of HER2-positive breast cancer cells were profiled for mutational status and also for anti-proliferative response to refametinib alone and in combination with the PI3K inhibitor (PI3Ki) copanlisib and the HER2-targeted therapies trastuzumab and lapatinib. Reverse phase protein array (RPPA) was used to determine the effect of refametinib alone and in combination with PI3Ki and HER2-inhibitors on expression and phosphorylation of proteins in the PI3K/AKT and MEK/MAPK pathways. We validated our proteomic in vitro findings by utilising RPPA analysis of patients who received either trastuzumab, lapatinib or the combination of both drugs in the NCT00524303/LPT109096 clinical trial.

Results: Refametinib has anti-proliferative effects when used alone in 2/3 parental HER2-positive breast cancer cell lines (HCC1954, BT474), along with 3 models of these 2 cell lines with acquired trastuzumab or lapatinib resistance (6 cell lines tested). Refametinib treatment led to complete inhibition of MAPK signalling. In HCC1954, the most refametinib-sensitive cell line (IC50= 397 nM), lapatinib treatment inhibits phosphorylation of MEK and MAPK but activates AKT phosphorylation, in contrast to the other 2 parental cell lines tested (BT474-P, SKBR3-P), suggesting that HER2 may directly activate MEK/MAPK and not PI3K/AKT in HCC1954 cells but not in the other 2 cell lines, perhaps explaining the refametinib-sensitivity of this cell line. Using RPPA data from patients who received either trastuzumab, lapatinib or the combination of both drugs together with chemotherapy in the NCT00524303 clinical trial, we found that 18% (n=38) of tumours had decreased MAPK and increased AKT phosphorylation 14 days after treatment with HER2-targeted therapies. The combination of MEK inhibition (MEKi) with refametinib and copanlisib led to synergistic inhibition of growth in 4/6 cell lines tested (CI @ED75 = 0.39-0.75), whilst the combinations of lapatinib and refametinib led to synergistic inhibition of growth in 3/6 cell lines (CI @ED75 = 0.39-0.80).

Conclusion: Refametinib alone or in combination with copanlisib or lapatinib could represent an improved treatment strategy for some patients with HER2-positive breast cancer, and should be considered for clinical trial evaluation. The direct down-regulation of MEK/MAPK but not AKT signalling by HER2 inhibition (e.g. by lapatinib or trastuzumab), which we demonstrate occurs in 18% of HER2-positive breast cancers may serve as a potential biomarker of responsiveness to the MEK inhibitor refametinib.

Keywords: HER2-positive breast cancer; MEK inhibitor; acquired resistance to HER2-targeted therapies; reverse phase protein array.

Conflict of interest statement

CONFLICTS OF INTEREST The authors state that they have no conflicts of interest in relation to this article or the funding bodies.

Figures

Figure 1. Differential expression of MEK1 as determined by RPPA in a panel of breast cancer cell lines (n=28) dependant on their sensitivity to the MEK inhibitors tremitinib and PD-0325901

Standard deviations are calculated from analysis of MEK expression in the sensitive (n=15) versus the resistant (n=12) cell lines analysed on the same RPPA slide. ‘*’ indicates a significant p value of <0.05 as determined by the students t-test.

Figure 2

(A) RPPA analysis displaying the fold-change in protein expression or phosphorylation relative to untreated controls in cell lines (HCC1954-P, -L) treated with 300nM refametinib (MEKi). (B) RPPA analysis displaying the basal levels of EGFR and ERBB3 protein expression and phosphorylation in the HCC1954-P and HCC1954-L cells. Standard deviations are calculated from triplicate independent protein samples analysed on the same RPPA slide. ‘*’ indicates proteins which have a change of signal intensity greater than 1.2 fold and a p-value of <0.05 as determined by the students t-test.

Figure 3

(A) RPPA analysis displaying the fold-change in protein expression or phosphorylation relative to control treated cells in cell lines (HCC1954-P, BT474-P and SKBR3-P) treated with 150nM lapatinib for 30 minutes (Lap30min). (B) Representative figure demonstrating hypothesised inhibition of MAPK/ERK signalling in HCC1954 and SKBR3 cells as a result of lapatinib treatment. Standard deviations are calculated from triplicate independent protein samples analysed on the same RPPA slide. ‘*’ indicates proteins which have a change of signal intensity of greater than 1.2 fold and a p value of <0.05 as determined by the students t-test.

Figure 4

RPPA analysis of the NCT00524303 clinical trial looking at changes in AKT and MAPK signalling pathways in tumours in response to treatment with either (A) trastuzumab, (B) lapatinib or (C) a combination of both. Patient samples where AKT (S473) is increased by >20% and MAPK(T202/Y204) is decreased by >20% are indicated in red.

Figure 5. Efficacy of refametinib (MEKi) (-□-), copanlisib (PI3Ki) (-◊-) and a combination of refametinib and copanlisib (--∆--) in a panel of HER2-positive breast cancer cell lines, including parental cells (-P) and those with acquired resistance to either trastuzumab (-T or -Res) or lapatinib (-L)

Error bars are representative of standard deviations across triplicate independent experiments. The ratio of refametinib:copanlisib in this assay was fixed at either 20:1 or 200:1.

Figure 6. RPPA analysis displaying the fold-change in protein expression or phosphorylation relative to control treated cells in cell lines treated with either 300nM refametinib (MEKi) or 15nM copanlisib (PI3Ki) alone or in combination (MEKi - 300nM: PI3Ki - 15nM) in HCC1954-P (parental) and -L (lapatinib resistant) cells

Standard deviations are calculated from at least triplicate biologically independent protein samples analysed on the same RPPA slide. ‘*’ indicates proteins which have a change of signal intensity of greater than 1.2 fold and a p-value of

Figure 7. RPPA analysis displaying the fold-change in protein expression or phosphorylation relative to control treated cells in cell lines treated with either 300nM refametinib (MEKi) or Lapatinib (HCC1954-P - 150nM: HCC1954-L - 500nM) alone or in combination in HCC1954-P and -L cells

Standard deviations are calculated from at least triplicate biologically independent protein samples analysed on the same RPPA slide. ‘*’ indicates proteins which have a change of signal intensity of greater than 1.2 fold and a p-value of

Figure 8. Efficacy of refametinib (MEKi) (-□-), lapatinib (-◊-) and a combination of refametinib and lapatinib (--∆--) in a panel of HER2-positive breast cancer cell lines, including parental cell lines(-P) and matched cells with acquired resistance to either trastuzumab (-T or -Res) or lapatinib (-L)

Error bars are representative of standard deviations across triplicate independant experiments. The ratio of refametinib:lapatinib in this assay was fixed at 2:1.