MEK1 and MEK2 inhibitors

来源：榕意旅游网

REVIEWS

MEK1 and MEK2 inhibitors and cancer therapy:

the long and winding road

Christopher J. Caunt1, Matthew J. Sale2, Paul D. Smith3 and Simon J. Cook2

Abstract | The role of the ERK signalling pathway in cancer is thought to be most prominent in tumours in which mutations in the receptor tyrosine kinases RAS, BRAF, CRAF, MEK1 or MEK2 drive growth factor-independent ERK1 and ERK2 activation and thence inappropriate cell proliferation and survival. New drugs that inhibit RAF or MEK1 and MEK2 have recently been approved or are currently undergoing late-stage clinical evaluation. In this Review, we consider the ERK pathway, focusing particularly on the role of MEK1 and MEK2, the ‘gatekeepers’ of ERK1/2 activity. We discuss their validation as drug targets, the merits of targeting MEK1 and MEK2 versus BRAF and the mechanisms of action of different inhibitors of MEK1 and MEK2. We also consider how some of the systems-level properties

(intrapathway regulatory loops and wider signalling network connections) of the ERK pathway present a challenge for the success of MEK1 and MEK2 inhibitors, discuss mechanisms of resistance to these inhibitors, and review their clinical progress.

A cocktail of two protein kinase inhibitors, one inhibiting MEK1 and MEK2, and the other inhibiting glycogen synthase kinase 3 (GSK3).

The authors dedicate this article to Prof. Chris Marshall, FRS (1949–2015), an inspirational scientist whose work contributed enormously to our understanding of the RAS-regulated RAF–MEK–ERK pathway.

The ERK signalling pathway is activated by an array of receptor types, including receptor tyrosine kinases (RTKs), G protein-coupled receptors and cytokine recep-tors, and the core components of this pathway are now well known1,2. Activated RTKs recruit adaptor proteins and guanine nucleotide exchange factors (GEFs; such as SOS) to activate the HRAS, KRAS or NRAS GTPases at the inner leaflet of the plasma membrane (FIG. 1). Once activated, GTP-bound RAS (RAS–GTP) drives the for-mation of high-activity homodimers or heterodimers of the RAF protein kinases (ARAF, BRAF or CRAF), which directly phosphorylate and activate MEK1 and MEK2 (also known as MAPKK1 and MAPKK2). MEK1 and MEK2 are dual-specificity kinases that activate ERK1 and ERK2 by phosphorylating them at conserved threonine and tyrosine residues in the T-E-Y motif found in their activation loop. Hundreds of proteins have been defined as ERK1 and ERK2 substrates and ERK-interacting partners1,3; these include other pro-tein kinases and transcription factors (such as ETS and the activator protein 1 complex (AP1)), which regulate

Department of Biology and

Biochemistry, University of Bath, Claverton Down, Bath BA2 7AY, UK.2

Signalling Laboratory, The Babraham Institute,

Babraham Research Campus, Cambridge CB22 3AT, UK.3

AstraZeneca, Oncology iMed, Cancer Biosciences, Cancer Research UK, Li Ka Shing Centre, Cambridge Institute, Robinson Way, Cambridge CB2 0RE, UK.Correspondence to P.D.S. and S.J.C.

e-mails: paul.d.smith@astrazeneca.com;

simon.cook@babraham.ac.ukdoi:10.1038/nrc4000

the expression of immediate- and delayed-early genes such as the D-type cyclins to promote G1/S progression in the cell cycle4. ERK1 and ERK2 can also regulate cell survival by phosphorylating members of the apoptosis regulating BCL-2 protein family at the mitochondria5. ERK1/2 signalling regulates processes that are crucial for normal development, including cell proliferation, dif-ferentiation, survival and cell motility; indeed, germline deletion of some components of the ERK pathway causes embryonic lethality6, and a MEK1/2 inhibitor (MEKi) forms part of the 2i protocol that maintains embryonic stem cell pluripotency7. The same cellular processes are deregulated in cancer and represent some of the key hall-marks and driving characteristics of the cancer cell8,9. Many human cancers contain activating mutations in genes encoding RTKs, RAS, BRAF, CRAF, MEK1 or MEK2, which act as driving oncogenes; consequently, many cancers exhibit deregulated activation of, and an enhanced dependency on, ERK1/2 signalling.

The discovery of the core components of the RAS–ERK pathway2 kick-started a protein kinase drug dis-covery effort that continues today10–12. The first ERK pathway inhibitor to be discovered, PD98059, was reported 20 years ago13 and was shown to act inde-pendently of ATP as an apparent allosteric inhibitor of MEK1 and MEK2. Since then, MEKis have proved to be

VOLUME 15 | OCTOBER 2015 | 577NATURE REVIEWS | CANCER

REVIEWSAllosteric inhibitorA small molecule that inhibits the activity of an enzyme by binding to a regulatory site that is distinct from the active or catalytic site.

invaluable research tools, underpinning our knowledge of ERK1 and ERK2 biology and validating MEK1 and MEK2 as cancer drug targets14,15. As they are not ATP-competitive, it was correctly anticipated that MEKis would be more selective than conventional kinase inhibi-tors and widely hoped that this would translate into rapid clinical success. In fact, it took a further 18 years until trametinib16 became the first MEKi to receive regulatory approval. In the interim, the identification of activating BRAF mutations (most notably BRAFT1799A encoding BRAF-V600E) in melanoma, thyroid and colorectal cancer (CRC) in 2002 (REF. 17) galvanized the field and focused efforts on developing BRAF inhibitors (BRAFis) a

EGFR or FGFRSOSRASGTPRAFRAFSOSthat proved to be BRAF-V600E selective, culminating in 2008 in the description and subsequent approval of vemurafenib18,19 and dabrafenib20,21.

In this Review, we consider the fundamental aspects of the ERK pathway from a MEK perspective, and the evidence that supports MEK1 and MEK2 as drug targets in cancer. We describe the MEKis, their unique modes of action and innate resistance mechanisms, and how tumour cells that are sensitive to MEKis can adapt and acquire resistance. We also discuss how knowledge of MEKi resistance has informed combination strategies and review the clinical experiences of MEKis, including successes and lessons learned.

RASGTPPRAFRAFKSR1SOSPPRASGTP–PRAFRAFPKSR1––PMEK1/2–SPRYMEK1/2MEK1/2––PDUSP6ERK1/2CytoplasmNucleusERK1/2PPETSERK1/2ERK1/2ERK1/2PPETSERK1/2PPETSDUSP5Figure 1 | Scaffolds and feedback controls underpin the normal functioning of the ERK1/2 pathway. a | A simplified

| Cancerlinear representation of the RAS-regulated RAF–MEK–ERK signalling cascade. Activated growth factor receptors (for

example, epidermal growth factor receptor (EGFR) and fibroblast growth factor receptor (FGFR)) recruit the guanine nucleotide exchange factor (GEF) SOS to promote the release of GDP from RAS, allowing GTP to bind. Active RAS

(RAS–GTP) drives the formation of active homodimers or heterodimers of the RAF protein kinases (ARAF, BRAF or CRAF), which directly phosphorylate and activate MEK1 and MEK2 (MEK1/2). MEK1/2 are dual-specificity protein kinases that

phosphorylate ERK1 and ERK2 (ERK1/2) on a conserved TEY motif to activate them. ERK1/2 substrates include transcription factors of the ETS family; in this way the pathway links activated receptors at the plasma membrane to changes in gene expression in the nucleus. The concentration of the core components typically increases down the pathway such that [RAF] < [MEK] < [ERK]. This signal amplification allows low-level receptor occupancy to elicit meaningful ERK signals throughout the cell166–168. b | Activation of the ERK pathway is critically dependent on scaffold proteins. Scaffolds serve several roles, including increasing the efficiency of interactions between the enzyme and substrate at each step and insulating pathway components against inputs from other parallel pathways to ensure signal fidelity22,169,170. For example, the kinase suppressor of RAS 1 (KSR1) scaffold assembles RAF, MEK1/2 and ERK1/2 to increase signalling efficiency; acts allosterically to activate the RAF kinase domain171; controls the subcellular location of the pathway; and insulates it from other pathways. Scaffolds tend to make signal transmission more efficient but limit amplification. c | The ERK pathway is extensively regulated by homeostatic negative feedback controls that fine-tune pathway output166. Rapid and direct feedback mechanisms involve the phosphorylation of MEK1, CRAF, BRAF, KSR1, SOS and some receptor tyrosine kinases (RTKs) by ERK and downstream kinases (such as RSK) to inhibit signal propagation23,101. Notably, ERK can phosphorylate CRAF and BRAF to inhibit MEK phosphorylation172–174, and MEK1 to inhibit ERK phosphorylation41,42. Loss of ERK activity (for example, by treatment with a MEK inhibitor (MEKi)) collapses these feedback loops and reactivates MEK and ERK; this confers ‘robustness’, allowing the pathway to adapt to perturbations166,175 and explains the ERK reactivation that is observed in tumour cells with wild-type BRAF75. The slower de novo expression of Sprouty (SPRY) proteins and the dual-specificity phosphatases (DUSPs) also regulates pathway output. SPRY proteins inhibit ERK signalling at the level of RTKs, SOS and by interfering with the RAF catalytic domain176. The DUSPs inactivate ERK by dephosphorylating the pT-E-pY motif. ERK-driven expression of DUSPs provides homologous intrapathway feedback to dampen pathway activation. Additionally, different DUSPs function in different locations, with DUSP5 residing in the nucleus and DUSP6 in the cytoplasm, allowing differential regulation of ERK output in these different locations24.

www.nature.com/reviews/cancerREVIEWS

The role of MEK in the ERK pathway

The ERK pathway is frequently represented as a lin-ear RAS–RAF–MEK–ERK signalling cascade (FIG. 1a) but this ignores the non-enzymatic components of the pathway and the layers of feed-forward and feed-back regulation that are vital for the role of the ERK pathway in information processing. Understanding how the pathway is activated by different input stim-uli to elicit different responses (such as proliferation or differentiation) and how the pathway responds and adapts to selective inhibition (for example, with MEKis) requires some appreciation of these systems-level features. First, activation of the ERK pathway depends on scaffold proteins such as kinase suppres-sor of RAS 1 (KSR1), which increase the efficiency of the inter actions between the enzyme and substrate at each step22 (FIG. 1b). Second, ERK1/2 signalling is criti-cally regulated by homeostatic feedback controls23 that include the direct phosphorylation of upstream com-ponents and increased expression of Sprouty (SPRY) proteins and dual-specificity phosphatases (DUSPs); the DUSPs inactivate ERK1 and ERK2 by dephosphory-lating the pT-E-pY motif24 (FIG. 1c). The importance of

DUSP feedback in restraining the oncogenic potential of MEK–ERK signalling is exemplified by the fre-quent loss of the ERK-specific cytoplasmic phos-phatase DUSP6 in epidermal growth factor receptor (EGFR)- and KRAS-driven non-small-cell lung cancers (NSCLCs)25 and the demonstration that loss of DUSP5 (the nuclear counterpart of DUSP6) in mouse models accelerates HRAS-driven skin cancer26. These negative feedback loops have additionally emerged as key deter-minants of rapid pathway adaptation and long-term acquired resistance to MEKis27,28. When taking into account scaffolding proteins and feedback loops, the canonical ERK pathway diagram looks more complex (FIG. 1c) but even this complexity fails to consider the position of the ERK pathway within wider signalling net-works (FIG. 2). MEK1 and MEK2 are the only activators of ERK1 and ERK2 and serve an entirely unique role as critical ‘ERK1 and ERK2 gatekeeper’ kinases, processing inputs from multiple upstream kinases. Indeed, a RAS- or RAF-centred view ignores the fact that RAF proteins are only a subset of the ‘MEK1 and MEK2 activators’ in cells (FIG. 2). Multiple MAP kinase kinase kinases (MAP3Ks) can activate MEK1 and MEK2, and some of

SOSIntegrinsRACPAKRASGTPMEKK1MEKK3MEKK2MAP3K8MLK2MLK4MLK1MLK3ARAFBRAFCRAFPPPMEK1/2MEK5ERK1/2PPBIMPPMCL1PPMYCPPRSKPPMSKCCND1PPMNKERK5MKK3 orMKK6p38MKK4 orMKK7JNK1/2IKKIκBCytoplasmPPPPFOSETSPPMEF2PPMEF2PPCHOPPPJUNNucleusNF-κBNature Reviews |The CancerFigure 2 | MEK1 and MEK2 are the key ‘gatekeepers’ for ERK1 and ERK2 in a wider signalling network.

canonical pathway for ERK activation (RAS–RAF–MEK–ERK) is shown on the left. ERK1 and ERK2 (ERK1/2) substrates include transcription factors (FOS, ETS and MYC) and other protein kinases (RSK, MNK and MSK), thereby controlling the transcription and translation of genes that promote cell cycle progression such as CCND1 (which encodes cyclin D1); other ERK substrates include regulators of apoptosis (BIM and MCL1). One key network feature is the convergence of signalling at the level of MEK1 and MEK2 (MEK1/2). Although ARAF, BRAF and CRAF are the best-studied MEK activators, a number of other MAP kinase kinase kinases (MAP3Ks; show on the right) can also fulfil this role, including MEKK1 (also known as MAP3K1), MEKK3 (also known as MAP3K3), MAP3K8 (also known as COT) and the mixed-lineage kinases

(MLK1–4; also known as MAP3K9, MAP3K10, MAP3K11 and KIAA1804)31–35. These ‘alternative’ MEK1/2 activators can also promote the activation of ERK5, p38, JUN N-terminal kinase (JNK) and nuclear factor-κB (NF-κB) (through IκB kinase or IKK) via their relevant upstream activating kinases (MEK5, MKK3 or MKK6, MKK4 or MKK7 or IκB kinase, respectively) to regulate transcription factors such as MEF2, CHOP, JUN or NF-κB31–35. This reflects a key feature of ERK pathway

architecture: receptor tyrosine kinases (RTKs)177, RAS GTPases2 and MAP3Ks31–35 are relatively promiscuous, activating two or more parallel signalling cascades; even RAF proteins may regulate non-MEK targets through scaffold functions178,179. By contrast, MEK1/2 are exquisitely specific activators of ERK1/2. Because ERK1/2 have hundreds of binding partners and substrates throughout the cell1,3, this represents an astonishing achievement in signal processing through MEK1/2 and underscores their role as key ‘gatekeepers’ of ERK signalling.

NATURE REVIEWS | CANCER

VOLUME 15 | OCTOBER 2015 | 579REVIEWSregion that stabilizes an inactive kinase conformation. The extreme C terminus contains the docking site for upstream activating MAP3Ks. Like many other kinases,

RAF andERKPAKthe MEK kinase domain consists of a small N-terminal MAP3Kslobe and a larger C-terminal lobe; conserved regions that are involved in ATP binding and hydrolysis, substrate Ser218Ser222Thr292Ser298Lys97binding, and phosphate transfer are found at the inter-PPPPface between these lobes. Activation of MEK1 and MEK2

MEK11DDNESNRRKinase catalytic domainALPRDDVD393requires conformational rearrangement of a C-helix in the N-lobe and the activation loop in the C-lobe to Ser222Ser226Lys101PPallow the correct alignment of ATP and substrate. This

rearrangement is caused by RAF- or MAP3K-mediated

MEK21DDNESNRRKinase catalytic domainALPRDDVD400phosphorylation of Ser218 and Ser222 within the MEK1 activation loop (Ser222 and Ser226 in MEK2)38,39.

Figure 3 | Linear representation of the key functional domains of the human MEK1

MEK1 is also regulated by the phosphorylation | Cancerand MEK2 proteins. Human MEK1 and MEK2 encode protein kinases of 393 amino

acids and 400 amino acids, respectively. The docking domain (DD) for ERK1 and ERK2, the of additional sites that are clustered within the kinase

domain. In addition to RTKs and RAS, MEK1 integrates nuclear export sequence (NES) and the MAP kinase kinase kinase (MAP3K) docking

signals from integrins via the RAC-dependent phos-domain (domain of versatile docking (DVD)) are shown in red and the amino-terminal

negative regulatory region (NRR) domain is shown in green. The kinase catalytic domain is phorylation of MEK1 in Ser298, which is catalysed by shown in blue and includes the highly conserved catalytic lysine residue (Lys97 or Lys101), p21-activated kinase (PAK1)40,41. MEK1 and MEK2 form the activation loop (AL), with sites of activating phosphorylation by RAF and other stable heterodimers in vivo that are unable to assemble MAP3Ks, and the proline-rich domain (PRD) that in MEK1 includes Thr292, a site of when MEK1 is phosphorylated by ERK1 and ERK2 on negative feedback phosphorylation by ERK1 and ERK2, and Ser298, a site of Thr292; this inability to form dimers decreases MEK1 phosphorylation by p21-activated kinase (PAK). Most gain-of-function mutations in

and MEK2 kinase activity as part of a negative feedback

MEK1 and MEK2 that are found in cancer or cardio-facio-cutaneous (CFC) syndrome

loop41,42. Notably, Thr292 is absent from MEK2, adding

cluster in the NRR or the N-terminal lobe of the kinase domain (shaded in purple). Figure

to a body of evidence that MEK1 and MEK2 have non-adapted: from REF. 36, Elsevier; from REF. 37, Springer; and from Bromberg-White, J. L.,

redundant roles, despite their high degree of homology Andersen, N. J. & Duesbery, N. S. MEK genomics in development and disease. Briefings in

and identical substrate specificity. For example, Mek2−/− Functional Genomics, 2012, 11, 4, 300–310, by permission of Oxford University Press.

mice apparently develop normally43, whereas Mek1 knockout causes embryonic death at embryonic day 10.5

these MAP3Ks are mutated in cancer and can confer (E10.5) owing to placental defects44.resistance to BRAFis30,31. Some of these MAP3Ks also

activate the JUN N-terminal kinase (JNK), p38, ERK5 MEK1 and MEK2 in cancer

and nuclear factor-κB (NF-κB) signalling pathways31–35, The importance of MEK1 and MEK2 in cancer first so that activation of the ERK pathway proceeds in con-emerged with the recognition of their strategic position cert with these other pathways (FIG. 2). Thus, specific in the RAS–RAF–MEK–ERK pathway and the demon-inhibition of MEK1 and MEK2 by allosteric inhibitors stration that activating mutations in the cDNAs encoding may cause substantial qualitative changes to signalling MEK1 and MEK2 that mimicked activation loop phos-networks. The extent to which these effects of MEKi phorylation could transform cells45,46. Subsequently, one are therapeutically desirable or contribute to toxicity of the earliest allosteric MEKis, PD184352 (also known in normal tissue is unclear. However, such qualitative as CI-1040), was shown to inhibit tumour cell prolif-changes in signalling are less likely to be observed in cells eration in vitro and to inhibit tumour growth in vivo14. that harbour BRAF-V600E and that are treated with a Since then, an array of studies have demonstrated that BRAFi, which leaves MEK–ERK activation by other various MEKis block tumour cell growth both in vitro MAP3Ks intact (see Supplementary information S1 and in vivo, underscoring the broad level of depend-(figure)). Clearly, understanding the biochemistry and ency of cancer cells on MEK1 and MEK2 in preclinical cell biology of MEK1 and MEK2 in the context of such models10–12,15. Such ‘MEK addiction’ seems to be strong-wider signalling networks is essential to understand their est in tumours that harbour BRAFV600E (REF. 47), which role in oncogenic signalling and to interpret the effects is consistent with the transforming effects of this onco-of MEKis.gene being mediated via the activation of MEK–ERK.

However, a considerable number of tumour cells that

Functional domains of MEK1 and MEK2. The second-are driven by RAS mutations are also sensitive to MEK1 ary structure of MEK1 and MEK2 (FIG. 3) comprises and MEK2 inhibition in vitro and in vivo47,48. This high-an amino-terminal sequence, a central protein kinase lights a key difference between MEKis and BRAFis. The domain (residues 68–361 in MEK1 and 72–367 in anti-proliferative effects of BRAFis are confined to cells MEK2) that contains the kinase activation loop and a that express BRAF-V600E or similar activating muta-MEK addiction

How dependent on MEK1/2 proline-rich segment, as well as a short carboxy-terminal tions in BRAF49. However, in the presence of RAS–GTP, activity a tumour cell is for sequence6,36,37. The N-terminal region contains an ERK1 BRAFis, such as vemurafenib, promote paradoxical acti-survival and proliferation; it can

and ERK2 docking site and a strong nuclear export vation of ERK1 and ERK2. This is a consequence of the broadly be measured by how

sequence that controls cytoplasmic MEK1 and MEK2 drug-induced allosteric transactivation of one BRAF sensitive a tumour cell is to a

MEK inhibitor.localization and overlaps with a negative regulatory molecule within RAS-induced RAF dimers and/or the

CFC and cancermutationsReceptorsand RASIntegrinswww.nature.com/reviews/cancerREVIEWS

prevention of RAF-inhibitory autophosphorylation50–53. In the clinic, this paradoxical ERK activation results in a range of secondary cutaneous lesions, including pap-illomas, squamous cell carcinomas, keratoacanthomas and basal cell carcinomas that profoundly limits the use of vemurafenib in the treatment of tumours that express BRAF-V600E21.

The first demonstration of activating mutations in MEK1 or MEK2 came from cardio-facio-cutaneous (CFC) syndrome, a genetic disorder that is caused by the aberrant activation of ERK1 and ERK2 during develop-ment. The first report of an amino acid-altering MEK mutation (encoding MEK2-P298L) was in a lung can-cer cell line in 1997 but the functional consequences were not defined55. Activating mutations in MEK1 or MEK2 were first reported in ovarian cancer cell lines in 2007 (REF. 56); since then, gain-of-function mutations in MEK1 or MEK2 have been reported in melanoma, CRC and lung cancer57–60. Most of these mutations cluster together with mutations that are found in CFC syndrome in either the N-terminal negative regulatory region or the ATP-binding region of the N-terminal lobe (FIG. 3). Notably, activating MEK1 mutations define a subset of

smoking-associated lung adenocarcinoma, which may account for up to 600 patients with lung cancer per year in the United States60. Although MEK1 and MEK2 mutations are rare in cancer as a whole, their existence provides an important level of validation for MEK1 and MEK2 as drug targets, and their incidence in lung can-cer defines a specific patient population that may ben-efit from MEKi therapy. In addition, MEK1 and MEK2 mutations have emerged as drivers of acquired drug resistance (see below). Finally, the deletion of both Mek1 and Mek2 in mice prevents the induction of NSCLC by endogenous KrasG12V (REF. 61). Interestingly, although CRAF is rarely mutated in human cancer its activity is strongly induced by mutant RAS proteins and Craf, but not Braf, is required for KrasG12V-driven lung cancers in mice61,62. Therefore, both CRAF and MEK1 and MEK2 are strongly validated drug targets in mouse models of lung cancer driven by mutant Kras.

Mechanism of action of MEK inhibitors

The first synthetic small-molecule inhibitor of MEK1 and MEK2 kinase activity to be discovered, PD98059, was reported in 1995 (REFS 13,63) (TABLE 1). Kinetic experiments

Table 1 | Properties and clinical progression of some widely used allosteric MEK1/2 inhibitorsMEK1/2 inhibitorPD098059U0126PD184352 (CI-1040)PD0325901Year Developer or reportedowner1995199819992004PfizerDuPontPfizerPfizerNovartis/Array BiopharmaAstraZeneca/Array BiopharmaBayer AGIn vitro IC50 for Ability to disrupt Clinical MEK1 (nM)*MEK phosphorylationprogression2000 ‡72‡17‡1‡1214‡WeakWeakWeakWeakWeakWeakPre-clinicalPre-clinicalPhase IIPhase IIPhase IIIPhase IIIT0.5 (hours or days)Not relevantNot relevant Refs13,63,73,7314,72,180||10,181||182–18568,73, 186114,187, 18875,79, 1,19019120.9+/−4.8 h7.8 h3.63–7.4 h5.33 hBinimetinib (MEK162, 2006ARRY-438162)Selumetinib (AZD6244, ARRY-142886)Refametinib (RDEA119, BAY 869766)CH4987655 (RO4987655)Pimasertib (AS703026, MSC1936369)TAK-733Trametinib (GSK1120212)CH5126766 (RO5126766)Cobimetinib (GDC-0973, XL518)GDC-06232007200919§WeakPhase II12 h2009Chugai Pharmaceutical CoMerck KGaA5.2ModeratePhase I4 h201052‡Not availablePhase II5 h20112011Takeda3.2‡WeakModeratePhase IApproved for BRAFV600E/K-mutant melanomaPhase I48–56 h~4 days192–19416,75,195GlaxoSmithKline0.72012Chugai Pharmaceutical CoGenentech (Roche)Genentech (Roche)160‡Strong60 h75,78,196201220134.25WeakStrongPhase IIIPhase I40 h4–10 h74,110, 19774,198*No unified protocol was used to generate the in vitro MEK1 IC50 values stated and so these should be treated as a rough guide to potency only. IC50 values shown for CH4987655, cobimetinib, GDC-0623 and trametinib were generated using methods in which MEK1 was activated after incubation with inhibitor. Apparent potency can differ greatly depending on whether constitutively active MEK1 (S218D/E S222D/E) is used (‡) or whether MEK1 is activated by RAF before inhibitor addition (§) as well as other experimental details. ||J. Sebolt-Leopold, personal communication (2015).

NATURE REVIEWS | CANCER

VOLUME 15 | OCTOBER 2015 | 581REVIEWSFeedback reliefERK1/2‑catalysed feedback phosphorylation and inhibition of RAF normally operates in the pathway but is relieved when MEK1/2 are inhibited and ERK1/2 activity declines, resulting in the activation of RAF and further MEK1/2 phosphorylation. Pathway activity also stimulates the expression of negative

regulators of pathway activity such as the ERK phosphatase DUSP6; expression of these negative regulators is reduced when the pathway is inhibited.

with PD98059 and U0126 (REF. ), another MEKi, dem-onstrated that both molecules shared a common binding site and inhibited MEK1 and MEK2 non-competitively with respect to their substrates Mg-ATP and ERK1 and ERK2. These early MEKis served as valuable research tools but their low potency, physical properties and cer-tain off-target effects65,66 stimulated the development of further generations of MEKi (TABLE 1).

PD184352 was the first MEKi to be evaluated in vivo and was shown to inhibit the growth of CRC tumour xenografts14. Crystal structures of MEK1 bound to ana-logues of PD184352 demonstrated the presence of a unique inhibitor-binding pocket that is separate from, but adjacent to, the Mg-ATP-binding site in both MEK1 and MEK2 (REF. 67); this region of MEK1 and MEK2 exhibits little homology to other protein kinases, thus providing an explanation for why this class of drugs is so specific68. Allosteric MEKis, such as PD184352, stabilize an inactive conformation of MEK1 and MEK2 in which helix C and the activation loop are displaced, resulting in the misalignment of catalytic residues and the pos-sible partial occlusion of the ERK1 and ERK2 activation loop binding site67–69. Although the clinical progression aEGFR or FGFRRASGTPRASof PD184352 was curtailed by poor biological availability and potency70, the past 10 years have seen the discov-ery of highly selective allosteric MEKis with far superior pharmacological and pharmaceutical properties (TABLE 1).Until recently, allosteric MEKis were thought to act through broadly equivalent mechanisms71; however, recent studies have changed this view. In cells with wild-type BRAF, including those with RAS mutations, ERK-dependent feedback phosphorylation of BRAF and CRAF inhibits the binding of both BRAF and CRAF to RAS–GTP and disrupts BRAF–CRAF heterodimers, thus reducing RAF signalling to MEK1 and MEK2 (FIG. 1c). As a result, the loss of ERK1 and ERK2 activity following MEK1 and MEK2 inhibition results in the dephosphory-lation and activation of CRAF (so-called feedback relief) and a CRAF-dependent increase in phosphorylation of the MEK1 and MEK2 activation loop and reactivation of ERK1 and ERK2 (FIG. 4). This has been observed for many MEKis, including PD098059, U0126, PD184352, PD0325901, selumetinib (also known as AZD6244 and ARRY-142886) and cobimetinib72–75. However, enhanced MEK1 and MEK2 phosphorylation in response to MEKis is not observed in BRAF-mutant cells,

cbGTPRASGTPRASGTPRASRAFRAFGTPRAFRAFRAFRAFRAFRAFPPMEK1/2PPMEK1/2• CH5126766• GDC-0623• TrametinibRAFRAFPPMEK1/2MEK1/2PPERK1/2• PD0325901• Selumetinib• CobimetinibPPERK1/2ERK1/2DUSP• Proliferation• SurvivalAdaptive resistanceFigure 4 | MEKis that inhibit MEK1/2 phosphorylation suppress the rebound in ERK1/2 activation that results | Cancerfrom relief of negative feedback. a | With the exception of tumour cells that harbour activating BRAF mutations, ERK1 and ERK2 (ERK1/2) signalling is subject to extensive ERK1/2-dependent negative feedback at multiple levels of the pathway, including RAF. RAS–GTP drives the formation of high activity homodimers or heterodimers of the RAF protein kinases (ARAF, BRAF and CRAF), whereas ERK-dependent phosphorylation of RAF proteins inhibits the binding of BRAF and CRAF to RAS–GTP and disrupts BRAF:CRAF heterodimers, thereby inhibiting phosphorylation of MEK. b | MEK1 and MEK2 (MEK1/2) inhibition with a MEK inhibitor (MEKi) blocks ERK1/2 activation and so relieves this negative feedback and consequently allows stronger activation of upstream pathway components, including RAS and RAF (represented by increased numbers of active RAS molecules and active RAF dimers). The majority of MEKis, such as PD0325901,

selumetinib and cobimetinib, do not disrupt the phosphorylation of the MEK1/2 activation loop sites and so treatment with these MEKis and concomitant relief of negative feedback typically results in the accumulation of phosphorylated MEK1/2. This is thought to explain the rebound in phosphorylated ERK1/2 and pathway output that is observed with these MEKis in various contexts, notably in tumour cells with mutant RAS74,75. c | A subset of newer MEKis, so-called

‘feedback busters’, including trametinib, CH5126766 and GDC-0623, disrupt the conformation of the MEK1/2 activation loop sites so that they can no longer be efficiently phosphorylated by RAF. Thus, although MEK inhibition with these MEKis is also expected to result in stronger activation of upstream pathway components, little or no increase in MEK1/2

phosphorylation or rebound in ERK1/2 activity is observed (right), translating into more durable pathway inhibition and superior efficacy in preclinical models75.

www.nature.com/reviews/cancerREVIEWS

Feedback buster

A term adopted by the field, although something of a

misnomer. All MEK inhibitors (MEKis) relieve ERK‑dependent negative feedback to RAF, resulting in RAF activation. Feedback buster MEKis mitigate some, but not all, consequences of feedback relief that arise when ERK is inhibited by interfering with the phosphorylation of MEK by RAF, thereby reducing rebound activation of the pathway. By contrast, conventional MEKis do not prevent MEK phosphorylation by RAF.

Phosphomimetic mutant

A phosphomimetic mutant of MEK1 exhibits constitutive (MAP3K‑independent) activation owing to acidic substitutions at Ser218 and Ser222 in the activation loop that mimic the negative charge of phosphorylation.

especially BRAF-V600E melanoma, principally because BRAF-V600E is fully active as a monomer and insensi-tive to ERK1- and ERK2-dependent inhibitory phospho-rylation and the inhibitory action of SPRY proteins, but also because the activity of RAS and CRAF is typically low in BRAF-V600E melanoma72,76,77. Intriguingly, some newer MEKis — including trametinib, CH5126766 and GDC-0623 — while inhibiting negative feedback mecha-nisms, mitigate the consequences of this by reducing the phosphorylationn of MEK by RAF74,78 (FIG. 4). This may explain why these inhibitors cause more durable sup-pression of ERK1 and ERK2 phosphorylation, cell pro-liferation and xenograft growth in the context of RAS mutations74,75. It is important to understand the under-lying mechanism of these feedback buster MEKis because KRAS-mutant tumours represent an important unmet clinical need and there is a growing interest in using MEKis as part of drug combinations in KRAS-mutant disease, including NSCLC.

Inhibition of activation loop phosphorylation. Differences in the mechanism of action may reflect distinct inter actions between the MEKi and the MEK1 and MEK2 activation loop residues. For example, GDC-0623 is proposed to form a stronger hydrogen bond inter action with Ser212 than other inhibitors, thereby constraining the activation loop and disrupting phosphorylation by RAF74. However, crys-tal structures of MEK1 bound to various allosteric MEKis — such as CH4987655, CH5126766, and PD184352- and PD0325901-like inhibitors — suggest that interaction with Ser212 is invariably crucial for the binding of allosteric MEKis to MEK1 and MEK2 (REFS 67,69,75,79). Instead, the ability to disrupt MEK1 and MEK2 phosphory lation correlates with the extent to which a MEKi coordinates with Asn221 and Ser222 to displace the activation loop75,79. Whereas CH5126766 binds both Asn221 and Ser222 and causes activation loop displacement, a near-identical enantiomer of PD0325901 does not coordinate Asn221 and Ser222 or displace the activation loop, which is con-sistent with the differing abilities of these inhibitors to disrupt MEK1 and MEK2 phosphorylation69,75. Whether these MEKis antagonize phosphorylation of MEK1 at both Ser218 and Ser222 (and of MEK2 at Ser222 and Ser226) is unknown; mass spectrometry suggests that trametinib prevents the phosphorylation of MEK1 at Ser218, but not at Ser222, and this monophosphorylated form of MEK1 has severely limited kinase activity compared with dual-phosphorylated MEK1 (REF. 16). Further structural analyses will determine whether, and how, distinct MEKis disrupt the phosphorylation of MEK1 and MEK2 and may guide future MEKi development.

Modulation of RAF–MEK1/2 complexes. MEKis also modulate the interaction between MEK1 or MEK2 and RAF, and although this may influence their ability to sup-press MEK1 and MEK2 activity, no simple correlation is apparent. For example, selumetinib and PD0325901 induce the binding of all three RAF kinases to MEK1 and MEK2, and this seems to attenuate the activity of these MEKis75. By contrast, GDC-0623 and CH5126766 also promote the association of RAF with MEK1 and MEK2,

but inhibit their phosphorylation74,75,78. Both trametinib and cobimetinib promote the dissociation of RAF–MEK1/2 complexes, but whereas trametinib antago-nizes MEK1 and MEK2 phosphorylation, cobimetinib does not74,75. Thus, although a firm causal relationship between MEKi activity and the association or disso-ciation of the RAF–MEK1/2 complex has not yet been established, the modulation of these complexes may be integral to their mode of action and may influence the depth and duration of pathway inhibition.

Feedback buster MEKis in the context of BRAFV600E mutations. Intriguingly, MEKis that disrupt MEK1 and MEK2 phosphorylation seem to have a relatively weaker affinity for dual-phosphorylated MEK1/2 than do the MEKis that permit the accumulation of phosphorylated MEK1/2. The binding of GDC-0623 to a constitutively active phosphomimetic mutant of MEK1 and the binding of trametinib to MEK1 that had been pre-phosphory-lated in vitro were both markedly weaker than binding to wild-type MEK1 and unphosphorylated MEK1, respec-tively16,74. By contrast, cobimetinib bound with similar affinity to wild-type and phosphomimetic MEK1, and suppressed ERK phosphorylation more effectively than GDC-0623 in cells expressing phosphomimetic MEK1 (REF. 74). Thus, for certain feedback buster MEKis, an equilibrium may exist between the inhibition of MEK1 and MEK2 activation loop phosphorylation and the attenuation of MEKi binding by the phosphorylated activation loop, with the balance being determined by pretreatment phospho-MEK1/2 levels. These observa-tions may be of particular relevance when constitutive MEK1 and MEK2 phosphorylation levels are high, such as in BRAF-V600E-positive cells; indeed, there is some support for this proposal from BRAFV600E-mutant xenograft models72. Clearly, careful consideration of the signalling context and particular properties of a MEKi will be required to achieve optimal clinical responses, especially in RAS-mutant tumours.

Resistance to MEKi and mitigation

Preclinical and clinical studies have led to the identifica-tion of various modes of innate, adaptive and acquired resistance to MEKis, many of which are druggable, allowing relevant combination strategies to be tested.

Intrinsic resistance through parallel oncogenic path-ways. Primary sensitivity to MEKi correlates with the decreased expression of cyclin D1 (CCND1), expres-sion of p27 (also known as KIP1) and cell cycle arrest. However, the deregulation of the cyclin-dependent kinases 4 and 6 (CDK4/6)–RB axis (such as through the amplification of CCND1 or CDK4 or the loss of CDK inhibitor 2A (CDKN2A)) is common in many cancers80 and can confer resistance to ERK pathway inhibitors81. Indeed, activating mutations in RAS, BRAF or MEK1 and MEK2 can co operate with CDKN2A loss in a variety of

82–84

tumours. Such results have led to the effective combi-nation of BRAFis or MEKis with inhibitors of CDK4/6 in preclinical models85,86. Other signalling pathways — such as PI3K, adenomatous polyposis coli (APC)–β-catenin,

VOLUME 15 | OCTOBER 2015 | 583NATURE REVIEWS | CANCER

REVIEWSEGFR or HER2HER3PDGFRβ, VEGFR or AXLFGFR2 or FGFR3RASRAFPPI3KRAFSTAT3MEKiCytoplasmNucleusMEK1/2ERK1/2• Proliferation• SurvivalERK1/2CtBPBETiERBB3NRG1MYCBETiPDGFRBVEGFR2AXL?BETiFGFR2FGFR3Inhibit geneexpressionFigure 5 | Adaptive kinome reprograming arising from MEK1/2 inhibition. Active ERK1 and ERK2 (ERK1/2) chronically

| Cancerinhibit signalling from an array of receptor tyrosine kinases (RTKs) by direct phosphorylation of receptors at inhibitory sites

(for example, ERK-mediated phosphorylation of epidermal growth factor receptor (EGFR) and HER2) or by repressing the transcription of the genes that encode RTKs and their cognate ligands, mediated by ERK1/2-dependent phosphorylation of transcriptional regulators such as CtBP and MYC (solid arrows). Inhibition of MEK1 and MEK2 (MEK1/2) collapses these feedback loops, resulting in rapid and sustained reactivation of multiple RTKs that can then sustain cell survival and proliferation by reactivation of RAF–MEK–ERK signalling or by activation of PI3K or signal transducer and activator of transcription 3 (STAT3)-dependent signalling28,96–99 (dashed arrows). Such kinome reprogramming validates clinical trials in which MEK inhibitors (MEKis) are being combined with various RTK inhibitors. However, MEKis frequently cause activation of multiple RTKs so an alternative, broader approach is to combine a MEKi with BET domain inhibitors (BETis), which act on key chromatin reader proteins to inhibit transcription100. FGFR, fibroblast growth factor receptor; NRG1, neuregulin 1; PDGFRβ, platelet-derived growth factor receptor-β; VEGFR, vascular endothelial growth factor receptor.

signal transducer and activator of transcription 3 (STAT3) and NF-κB — converge on the same cell cycle regulators and can drive primary MEKi resistance. For example, even in tumour cells that harbour BRAFV600E, strong PI3K-dependent signalling owing to mutations in PIK3CA or the loss of PTEN can maintain CCND1 levels in the presence of MEKis and can confer resist-ance to MEKis87,88. This can be overcome by combining MEKis with PI3K, mTOR or AKT1 (also known as PKB) inhibitors. In addition, the frequent increase in PKB/AKT activity following treatment with MEKis has prompted numerous studies with these drug combinations,90 and ongoing clinical trials91. STAT3 activation has been shown to promote MEKi resistance; combining a STAT3 inhibi-tor with selumetinib overcame resistance and promoted tumour cell death92,93. More recently, the Hippo pathway effector YAP1 has been shown to promote resistance to RAFi or MEKi therapy, and combined inhibition of YAP1 and MEKi was synthetic lethal in tumour cells94.

For example, tumour cell lines with BRAFV600E exhibited durable inhibition of ERK1 and ERK2 and were very sensitive to PD0325901, whereas tumour cell lines with mutations in KRAS were notably resistant to PD0325901 and exhibited strong ERK1 and ERK2 reactivation within 24–48 hours of drug administration75. Thus, hardwired homeostatic mechanisms dictate the efficacy of MEKi in tumour cells with wild-type BRAF, including those with mutations in RAS, and provide a clear rationale for dual pathway inhibition: RAFi plus MEKi to prevent MEK1 and MEK2 reactivation, or MEKi plus ERKi to prevent ERK1 and ERK2 reactivation. An alternative approach may be to use feedback buster MEKis with a dual mecha-nism — such as trametinib, CH5126766, GDC-0623 and G-573 — that inhibit MEK1 and MEK2 kinase activity and prevent MEK1/2 phosphorylation by RAF74,75 (FIG. 4). Interestingly, despite the existence of multiple non-RAF MAP3Ks that can activate MEK1 and MEK2 (includ-ing MAP3K8 (also known as COT), some MEKKs and MLKs31–35) few studies have assessed whether these alter-Loss of feedback inhibition and MEK–ERK reactiva-native MEK activators are inhibited by ERK-dependent tion. Although MEKis have shown great promise in feedback phosphorylation and thus might contribute to BRAF-V600E preclinical models16,47, as well as some MEK reactivation following MEK inhibition.clinical activity in BRAF-V600E melanoma95, their

more limited success in RAS-mutant tumours may well Adaptive kinome reprogramming. ERK1/2 signal-be due in large part to the loss of ERK1/2-dependent ling inhibits an array of RTKs such that MEK1 and feedback and reactivation of the MEK–ERK1 pathway. MEK2 inhibition elicits rapid RTK activation (FIG. 5).

www.nature.com/reviews/cancerREVIEWS

For example, ERK1- and ERK2-mediated phosphory-lation of EGFR and HER2 (also known as ERBB2) sup-presses signalling to HER3 (also known as ERBB3) so that MEKis promote the rapid reactivation of EGFR–HER3 and PI3K-dependent signalling96,97. Unbiased, non-targeted chemical proteomic analysis demonstrated the activation of multiple RTKs following MEK1 and MEK2 inhibition, which were otherwise repressed by the ERK1 and ERK2 target MYC28. In BRAF-mutant thyroid cancer cells, BRAFis or MEKis promoted ERK1 and ERK2 reac-tivation involving HER3 transcription and the autocrine secretion of neuregulin 1 (NRG1)98, and in KRAS-mutant NSCLC cells selumetinib promoted autocrine fibroblast growth factor (FGF) production and increased expression of FGF receptors (FGFRs), which conferred MEKi resist-ance via STAT3 activation99. These studies underscore the extent of such ‘kinome re-programming’ in response to MEKis and support ongoing trials that are combining MEKis with relevant RTK inhibitors. However, as multiple RTKs are often activated in cancer and because transcrip-tional programmes underpin kinome reprogramming, an alternative strategy is to combine MEKis with inhibitors of key chromatin reader proteins such as the bromodomain- containing protein 4 (BRD4). Certainly, JQ1, a BET domain inhibitor, synergizes well with tyrosine kinase inhibitors in acute myelogenous leukaemia100 (FIG. 5).Acquired resistance to MEKis. Various studies have now investigated acquired resistance to long-term MEKi exposure, both in tumour cell lines and in clinical sam-ples101. The first report of acquired resistance to MEKis was the identification of a gain-of-function mutation (MEK1P124L) in a metastatic focus of a patient with melanoma with BRAFV600E, who exhibited prolonged stable disease under treatment with selumetinib102. Subsequently, mutations in MEK1 and MEK2 have been found in CRC, melanoma and breast cancer cell lines (with driving mutations in BRAF or KRAS) that have been treated with various MEKis (selumetinib, trametinib and RO4927350), as well as in samples from patients undergoing MEKi therapy103–105. These ‘on target’ mutations (MEK1F129L and MEK2Q60P) activate MEK1/2 or cluster at the allosteric inhibitor-binding pocket and abrogate MEKi binding (for example, MEK1L115P, MEK1G128D/L215P, MEK1F129L, MEK1V211D and MEK2V215E), thereby maintaining ERK1 and ERK2 activity in the presence of MEKis (FIG. 6).

Studies have also reported acquired resistance to MEKis via the amplification of BRAFV600E. Two stud-ies in four CRC cell lines harbouring BRAFV600E dem-onstrated the amplification of the mutant BRAF allele; MEKi resistance was overcome by BRAF RNA interfer-ence (RNAi) or co-treatment with a RAF inhibitor106,107. BRAFV600E amplification has also been observed in mela-noma cell lines selected in trametinib105, indicating that BRAFV600E amplification is a common mecha nism of resistance irrespective of the MEKi used. Amplification of KRASG13D has also been described as a mechanism of MEKi resistance in CRC cells107, either in isolation or concurrent with an activating MEK1F129L mutation103, suggesting that these oncogenes may cooperate to confer

NATURE REVIEWS | CANCER

MEKi resistance. Thus, amplification of the upstream driving oncogene (BRAFV600E or KRASG13D) drives resist-ance to MEKi by activating a greater pool of MEK1 and MEK2 to maintain ERK1 and ERK2 activity (FIG. 6).Strikingly, most, if not all, mechanisms of MEKi resistance reinstate active ERK1 and ERK2 in the pres-ence of MEKis (FIG. 6), thus highlighting a strong hard-wired addiction to ERK1 and ERK2 signalling. This provides another opportunity for rational therapeutic intervention through dual pathway inhibition; MEKi plus ERKi in the case of MEK1 and MEK2 mutations103 and BRAFi plus MEKi in the case of BRAFV600E ampli-fication106,107. Indeed, the combination of dabrafenib (a BRAFi) and trametinib (a MEKi) has now received reg-ulatory approval in advanced-stage melanoma, in which it increased progression-free survival (compared with either agent alone) and reduced the incidence of cutane-ous lesions arising from paradoxical RAF activation in tissues with wild-type BRAF21,108. However, recent studies have revealed that acquired resistance may still arise in patients with BRAFV600E melanoma who are treated with dual BRAFi plus MEKi blockade through the acquisition of BRAFV600E ‘ultra’ amplification, MEK1 mutations and/or loss of the CDKN2A locus, as well as loss of the nuclear ERK-regulatory phosphatase DUSP4 (REF. 27), echo-ing the recurring theme of ERK pathway re-activation.

G13D

In the case of KRAS amplification, resistance is more complex; KRAS activates multiple effector pathways, suggesting that ERK pathway inhibition may need to be combined with AKT, PI3K or mTOR kinase inhibitors. Indeed, such combinations are currently being tested in preclinical disease models and clinical trials–91.

More recently, the first potent, selective ERKis have been described, including VTX11e, an ATP-competitive inhibitor of ERK1 and ERK2, and SCH772984, an ATP-competitive inhibitor of ERK1 and ERK2 that also pre-vents the activating phosphorylation of ERK1 and ERK2 by MEK1 and MEK2 (REFS 103,104). Of these selective ERKis, SCH772984 can resensitize tumour cells with acquired resistance to either BRAFis or MEKis, provid-ing proof-of-principle for combining ERK inhibitors with BRAFis or MEKis. SCH900353 (MK08353), an analogue of SCH772984, has now entered clinical trials, as have BVD-523, RG7842 and CC-90003. These new potent and selective ERK inhibitors add new weapons to the arsenal of ERK1/2 pathway-targeted drugs for use as first-line therapies, probably in combination, and for the treatment of ‘on-pathway’ resistance.

MEKis in the clinical setting

Pharmacokinetics and dose schedule. In addition to dif-ferences in the mechanism of inhibition, the pharmaco-kinetic properties of the MEKis that are currently in clinical development vary considerably with half-lives (T0.5) ranging from 5 hours to 4–5 days (TABLE 1). Despite this, there has been little sustained effort to optimize the dose and schedule of administration on the basis of bio-logical rationale rather than pragmatic choices based upon pharmacokinetics and tolerance109, and most clini-cal trials are predicated upon continuous daily dosing. The exception is cobimetinib, which is administered on

VOLUME 15 | OCTOBER 2015 | 585REVIEWSaMEK1 or MEK2 mutation prevents MEKibinding or activates MEK1 or MEK2bBRAF ampliﬁcation maintains orincreases pool of active MEK1 or MEK2cKRAS ampliﬁcation increases pool of active MEK1 or MEK2 but also activates other RAS eﬀectorsKRASKRASKRASKRASKRASPLCεCa2+PKCBRAFBRAFBRAFBRAFARAFBRAFCRAFPI3KRAL-GEFMEK1/2MEK1/2PPERK1/2MitigationCombine MEKiwith ERKiMEK1/2MEK1/2PPERK1/2MitigationCombine MEKiwith RAFiMEK1/2MEK1/2MEK1/2PPERK1/2MEK1/2PDK1RALPKBmTORMitigationCombine MEKi with RAFiand PI3Ki, PKBi or mTORiFigure 6 | Mechanisms of acquired resistance to MEK1/2 inhibitors (MEKis). Studies in preclinical models and analysis of clinical samples have revealed two basic mechanisms of acquired resistance to MEK1 and MEK2 (MEK1/2) inhibitors (MEKis; represented by a red X), and, in both cases, tumour cells adapt to maintain or re-activate ERK1 and ERK2 (ERK1/2) in the presence of the drug101, providing a rational basis for treating resistance. a | The emergence of mutations in MEK1 or MEK2 (indicated by a hexagon), provides an example of ‘on-target’ resistance to allosteric MEKis. Emergent MEK1/2 mutations confer resistance by reducing MEKi binding or enhancing intrinsic MEK1/2 activity102–105 and resistance can be overcome by combining MEKi with ERK1/2 inhibitors. b | Amplification of the upstream driving oncogene, BRAFV600E (REFS 106,107) or KRASG13D (REF. 107), has also emerged as a mechanism of acquired MEKi resistance. Selective amplification of the mutant BRAFV600E allele greatly | Cancerincreases the proportion of active MEK1/2, exceeding the drug-inhibited pool, to reactivate ERK1/2; in this case, resistance can be overcome by combining a MEKi with a RAF inhibitor (RAFi)106,107. c | Amplification of KRASG13D can also drive MEKi resistance by the same basic mechanism107 but provides a greater therapeutic challenge. KRASG13D can activate

multiple signalling pathways (such as PI3K, RAL, phospholipase Cε (PLCε), and so on) and even the combined inhibition of MEK1/2 and PI3K signalling fails to reverse the acquired resistance to selumetinib that is driven by amplification of KRASG13D (REF. 107). Hexagons indicate oncogenic

mutations, either those that are present as the primary driving oncogene and are amplified by MEKi selection (BRAFV600E, KRASG13D) or those that emerge upon selumetinib selection (for example, MEK1P124L and MEK1F129L).

a 2 weeks on, 1 week off schedule; a pragmatic choice that is based upon its T0.5 of ~40 hours110, its adverse event profile and evidence of melanoma re-growth using schedules with longer drug holidays111.

Pharmacodynamics. The main biomarker for monitor-ing pathway inhibition by MEKis, phosphorylated ERK1 and ERK2 (p-ERK), has been assessed in normal tissues (such as peripheral blood mononuclear cells) and tumour biopsy samples. In multi-tumour-type Phase I studies, selumetinib, trametinib, BAY 86–9766 and pimasertib all reduced levels of p-ERK1 and p-ERK2 in tumour tissue using immunohistochemistry assays, and in some cases have shown complete inhibition112–114. On this basis and on the basis of data from BRAF inhibitors in BRAFV600E melanoma18 the prevailing view is that 80% inhibition of ERK1 and ERK2 phosphorylation is the benchmark for clinically effective MEK1 and MEK2 inhibition. However, these measures of target inhibition are based almost entirely on single time-point paired biopsies and non-standardized assays; no study has formally evaluated a link between target inhibition and clinical response. In summary, clinical dose and schedule selection for MEKis is mostly based on establishing the maximum tolerated doses using continuous or chronic dosing regimens.Clinical activity as monotherapy. As with preclinical models, BRAF-V600E-mutant melanoma has proved to be the most responsive adult solid tumour to MEKi

Dacarbazine

A DNA‑alkylating agent that has commonly been used as a single agent in the treatment of metastatic melanoma.

mono therapy. Trametinib has gained regulatory approval for this indication following rapid clinical develop-ment from a single-arm Phase I trial to a randomized Phase III trial in which treatment with this MEKi yielded 4.8 months progression-free survival compared with 1.5 months for dacarbazine95. In addition, the licensing and supervisory authority of Switzerland has recently approved cobimetinib for use in combination with vemurafenib as a treatment for patients with advanced melanoma. Such is the pace of clinical development in advanced-stage melanoma that MEKi monotherapy in BRAF-V600E melanoma — which is relatively ineffective in patients who relapse following treatment with a first-generation BRAFi115 — is being superseded by BRAFi plus MEKi combinations, of which three have completed or are in clinical trials111,116,117. MEKis have been tested in multiple tumour types with a high incidence of BRAF or RAS mutations with mixed results. Some diseases seem notably refractory; for example, KRAS-mutant CRC118,119. However, others have shown sufficient activity to prompt Phase III trials; for example, MEK162 in NRAS-mutant advanced-stage melanoma120 and selumetinib in serous low-grade ovarian cancer121, a disease in which MEK162 and trametinib are currently undergoing assessment in randomized Phase III trials. Between these extremes, MEKis have shown clear but low response rates in most indications, including pancreatic cancer109,122, biliary tract cancer123, NSCLC124, uveal melanoma125 and acute myeloid leukaemia (AML)126,127, and were generally not

REVIEWS

Plexiform neurofibromatosis

A non‑cancerous

hyperproliferative disease of the nerve sheath that is driven by the loss of the NF1 tumour suppressor.

Neoadjuvant

A neoadjuvant therapy is

treatment given before primary therapy; in the context of this Review, selumetinib is

administered first as adjuvant therapy and continued until the radioactive iodine therapy has been administered.

considered to warrant further monotherapy trials. For example, trametinib failed to show superiority over docetaxel in a trial in KRAS-mutant-positive NSCLC128. One conclusion from monotherapy trials is that most KRAS-mutant tumours are not highly addicted to ERK signalling, at least to the degree that it can be inhibited by MEKi monotherapy at tolerated doses using chronic dosing schedules. Possible solutions include better patient selection, optimizing dosing schedules and combination therapy. For example, there is a growing appreciation that relief of feedback RAF phosphorylation and reactivation of MEK and ERK may limit the monotherapy activity of the MEKis that do not prevent the phosphorylation of MEK by RAF (discussed above). Indeed, it is entirely possible that such feedback relief and reactivation of MEK is masking the true extent of ‘MEK addiction’ in KRAS-mutant tumours. Thus, one obvious rational combination therapy that should be tested for KRAS-mutant tumours is intrapathway dual inhibition (MEKi plus RAFi or MEKi plus ERKi).

Although responses to MEKi monotherapy have been modest in adult solid tumours, emerging data from the use of selumetinib in paediatric diseases — such as low-grade paediatric glioma and plexiform neurofibromatosis (neurofibromatosis type 1 (NF1) with inoperable plexi-form neurofibromas) — show an encouraging and dura-ble response rate129,130. Both diseases are characterized by ubiquitous mutational activation of RAS–ERK signalling and it is tempting to speculate that as low-grade diseases these tumours are less able to adapt to MEKis either through feedback ‘relief’ or compensatory pathway acti-vation, perhaps owing to low somatic mutation rates, as reported recently for pilocytic astrocytoma131.

Clinical activity in drug combinations. Randomized Phase III trials to assess MEKi in combination with first-generation BRAFis in BRAF-V600E-mutant melanoma have recently reported a clear benefit of the combination not only in efficacy but also in tolerability compared with a BRAFi or MEKi alone111,116,117. This successful combi-nation paradigm is, however, an outlier due to the two drugs synergizing in the tumour while antagonizing each other in normal tissue. Although it is highly unlikely that this type of combination effect can be reproduced with other classes of drugs the result highlights the impor-tant characteristics of a successful drug combination: mechanistic synergy, appropriate patient population selection and good tolerance. It is worth noting that the combination of a MEKi and a first-generation BRAFi is markedly less active in BRAF-mutant CRC132 in which relief of feedback and consequent activation of EGFR signalling is thought to limit response, prompting triple combination trials including anti-EGFR mono clonal antibodies (mAbs) with early, encouraging signs of activity133. Beyond dual pathway inhibition in BRAF-mutant melanoma the furthest advanced MEKi com-bination is selumetinib and docetaxel, which showed strong preclinical activity134,135 and is now in Phase III trials in KRAS-mutant NSCLC following a small ran-domized Phase II study in which the combination showed a benefit over docetaxel alone in terms of overall

response rate and progression-free survival136. A simi-larly encouraging response rate for trametinib in com-bination with docetaxel in NSCLC137 indicates that this combination is robustly active in this disease; it remains to be seen to what extent the antitumour efficacy is compromised by the tolerability burden. Other combi-nations of MEKi and standard-of-care cytotoxic drugs have fared less well: trametinib and pimasertib failed to progress beyond Phase II trials in combination with gemcitabine in pancreatic cancer138,139 and selumetinib in combination with irinotecan in KRAS-mutant CRC showed only modest activity140. Combinations with novel targeted agents, principally targeting PI3K, AKT or mTORC1, have required dose modifications, and although responses have been reported (for example, in NSCLC, pancreatic cancer and low-grade ovarian cancer) they have not yet shown definitive, confirmed

91,141,142

combination activity in the clinic.

Finally, one MEKi combination concept with intrigu-ing activity that may come close to fulfilling the criteria of synergy, selected patient population and tolerance is the use of a MEKi to re-sensitize iodine-refractory differentiated thyroid cancer (DTC) to radioactive iodine (RAI) therapy. The predominant oncogenic pathway in DTC, RAS–ERK, drives the loss of thyroid differentiation-specific gene expression, including the sodium–iodine transporter that takes up RAI. The result — inadequate uptake of and response to RAI — can be reversed by MEK inhibition, and this has prompted a Phase III neoadjuvant trial in patients with DTC who are at high risk of early relapse143,144.

Adverse clinical events associated with MEKis. Because MEK1 and MEK2 are rarely mutated in human cancer there is little prospect of the target-related, genetic thera-peutic index for a MEKi that is evident with the first gen-eration of BRAFis in BRAF-V600E/K melanoma; thus, normal tissue toxicity will constrain clinical activity, as evidenced by the difference in response rate between MEKis and BRAFis in BRAF-mutant melanoma. For example, trametinib (a MEKi) and dabrafenib (a BRAFi) achieved overall response rates of 22% and 53%, respec-tively, as single agents145–147. With at least eight MEKis in clinical development, the adverse event profile is becom-ing clear and characteristic: acneform rash, diarrhoea and fatigue are prevalent and generally well managed, whereas a range of ocular toxic effects, oedema, creatine phosphokinase elevations and cardiac complications are rarer but more troublesome148. The range of ocular toxic effects and their management have been reviewed else-where in detail149, and although many of these may be self-limiting, some effects require dose discontinuation or dose modification. Therefore, in conclusion, the dose and target engagement for all MEKis is currently limited by normal tissue toxic effects.

Combinations of MEKis and first-generation BRAFis are better tolerated than the respective monotherapies because the paradoxical activation of ERK signalling by the BRAFi in BRAF wild-type tissue antagonizes the pathway inhibition by the MEKi, and the MEKi limits the paradoxical activation of ERK signalling.

VOLUME 15 | OCTOBER 2015 | 587NATURE REVIEWS | CANCER

REVIEWSRASnessRefers to a transcriptomic signature that reports activation of the RAS pathway and may predict responses to RAS pathway drugs.

In this way, MEKi-induced skin rash and BRAFi-induced squamous lesions are diminished in combi-nation21. Combining MEKis with standard-of-care cytotoxic chemotherapy or various targeted agents has invariably increased the toxicity burden for patients. In the case of selumetinib in combination with cyto-toxic chemotherapy it has been possible to maintain the recommended monotherapy dose, albeit with enhanced toxicity136,140,150. By contrast, combinations with targeted therapies have generally required dose reduction or dosing schedule alteration of one or both agents141,151; for example, selumetinib in combination with MK-2206, an allosteric AKT inhibitor, or the com-bination of trametinib and pan-Class I PI3K inhibitor buparlisib (BKM120) showed some signs of clinical activity but was compromised by dose and schedule modifications to accommodate the enhanced toxic effects that are associated with dual-pathway inhibi-tion91,152. In the absence of a strong and specific clini-cal efficacy signal it is possible that combination toxic effects will further limit the clinical development of MEK inhibitors and agents that target PI3K–AKT signalling. Tolerance to trametinib in combination with cytotoxic chemotherapy is more variable: it can result in non-tolerance when trametinib is combined with platinum doublet chemotherapy, and it requires growth factor support when combined with docetaxel, and dose reduction when combined with peme-trexed153, although it does not require dose reduction in combination with gemcitabine138.

Patient populations. Although preclinical studies indi-cate a preferential activity for MEKis in tumours carry-ing BRAF and RAS mutations47,48,1,155, clinical validation of this concept is lacking despite the demonstration of clinical activity in selected patient populations. Even the approval of trametinib in BRAFV600E/K-mutant mela-noma did not demonstrate that trametinib was pref-erentially active in BRAF-mutant disease, as patients who did not have mutations in BRAF were not included in randomized trials95,145. Across many disease types MEKis have induced clinical responses in tumours with activating mutations in KRAS, NRAS, MEK1, GNAQ and GNA11 (guanine nucleotide-binding protein sub unit αq and subunit α11) and inactivation of the RAS GTPase-activating protein NF1 as well as in tumours in which no pathway-related mutation has been detected109,118,120–122,125,127,144,145. Progress towards the validation of patient selection biomarkers will require more rigorous trials. Although small non randomized trials of trametinib in combination with either docet-axel or pemetrexed in advanced NSCLC127,128,156 indi-cate that KRAS mutation status does not condition the response rate or progression-free survival, the outcome of ongoing randomized studies in NSCLC with prospective KRAS mutation testing should help to demonstrate with greater certainty the degree to which overall clinical benefit, compared with stand-ard of care, is dependent upon KRAS mutational sta-tus (ClinicalTrials.gov identifiers NCT01750281 and NCT01933932).Future perspectives

The past decade of MEKis clinical development has thrown into sharp focus the issues that will need to be addressed if their potential is to be realized. The use of MEKis with a shorter T0.5 at higher intermittent doses should test the hypothesis that a higher degree of pathway inhibition (with an inhibitor alone or in combination) can improve anti-tumour efficacy, delay time to resistance and also main-tain a therapeutic margin by sparing normal tissue157. The advent of next-generation sequencing platforms and more sensitive and quantifiable means for testing transcriptome signatures (such as those that measure ERK pathway out-put and RASness)48,158 will provide a step change in the breadth and depth of molecular information around which to develop patient selection biomarkers. Alternative com-bination partners are currently progressing rapidly in clini-cal trials. The success of intrapathway combinations to treat BRAF-V600E/K melanoma in the clinic is mirrored by pre-clinical data on KRAS-mutant tumours combining MEKis with RAF dimer inhibiting BRAFis159 (WO2008120004) or with ERKis. Indeed, the entry of ERK1 and ERK2 inhibitors into clinical trials means that ‘on-target’ muta-tions in ERK1 and ERK2 should be anti cipated as possible

160

mechanisms of acquired resistance. Intrapathway com-binations will be more challenging to develop clinically owing to the probable added toxicity burden, although intermittent dosing may mitigate this added burden. MEK1 and MEK2 inhibition ultimately regulates the G1/S transition, and therefore efficacy may be compromised in tumours with lesions on the cyclin D–CDK4/6–RB pathway (for example, loss of CDKN2A)86,88,161. Indeed, early clinical data indicate that the combination of MEKis and CDK4/6is holds some promise162. As MEKis predomi-nantly exert cytostatic effects, identifying drug combina-tions that harness ‘MEK addiction’ to promote tumour cell synthetic lethality will prove increasingly important and hopefully fruitful. Indeed, this has already been achieved as a preclinical proof-of-principle with BH3 mimetics. MEKis modulate the apoptotic threshold by stabiliz-ing the pro-apoptotic protein BIM; in combination with BH3 mimetics that antagonize either BCL-2 or BCL-XL, BIM is liberated to neutralize anti-apoptotic BCL-2 fam-ily members such as MCL1, or to directly activate BAX, thus driving tumour cell death163,1 and thereby delaying

163

the onset of MEKi resistance. Most importantly, given the rapid recent advancement and success of tumour immu-notherapies, it is vital that the combination of MEKis with immune checkpoint inhibitory antibodies (for example, programmed cell death protein 1 ligand 1 (PDL1)-specific antibodies that antagonize the immune-suppressive effects of PDL1, thereby boosting tumour immunity) is evaluated. The complex effects of MEKis on the relationship between the tumour and the immune environment are formida-ble, comprising both positive (for example, tumour cell death, release of antigens, increased MHC expression and increased immune system priming) and negative effects (for example, inhibitory effects on T cell activation and pro-liferation)165. Optimization of dose and schedule and the identification of patient populations in which RAS–ERK signalling may exert an immune-suppressive effect will be the key to success.

REVIEWS

Yoon, S. & Seger, R. The extracellular signal-regulated kinase: multiple substrates regulate diverse cellular functions. Growth Factors 24, 21–44 (2006).

2. Karnoub, A. E. & Weinberg, R. A. Ras oncogenes: split

personalities. Nat. Rev. Mol. Cell Biol. 9, 517–531 (2008).

3. von Kriegsheim, A. et al. Cell fate decisions are

specified by the dynamic ERK interactome. Nat. Cell Biol. 11, 1458–14 (2009).

4. Meloche, S. & Pouysségur, J. The ERK1/2 mitogen-activated protein kinase pathway as a master

regulator of the G1- to S-phase transition. Oncogene 26, 3227–3239 (2007).

5. Balmanno, K. & Cook, S. J. Tumour cell survival

signalling by the ERK1/2 pathway. Cell Death Differ. 16, 368–377 (2009).

6. Bromberg-White, J. L., Andersen, N. J. &

Duesbery, N. S. MEK genomics in development and disease. Brief. Funct. Genom. 11, 300–310 (2012).7. Ying, Q. L. et al. The ground state of embryonic stem

cell self-renewal. Nature 453, 519–523 (2008).8. Hanahan, D. & Weinberg, R. A. The hallmarks of

cancer. Cell 100, 57–70 (2000).

9. Hanahan, D. & Weinberg, R. A. Hallmarks of cancer:

the next generation Cell 144, 6–674 (2011).10. Sebolt-Leopold, J. S. & Herrera, R. Targeting the

mitogen-activated protein kinase cascade to treat cancer. Nat. Rev. Cancer. 4, 937–947 (2004).11. Montagut, C. & Settleman, J. Targeting the RAF–

MEK–ERK pathway in cancer therapy. Cancer Lett. 283, 125–134 (2009).

12. Samatar, A. A. & Poulikakos, P. I. Targeting RAS–ERK

signalling in cancer: promises and challenges. Nat. Rev. Drug Discov. 13, 928–942 (2014).

13. Dudley, D. T., Pang, L., Decker, S. J., Bridges, A. J. &

Saltiel, A. R. A synthetic inhibitor of the mitogen-activated protein kinase cascade. Proc. Natl Acad. Sci. USA 92, 7686–76 (1995).

This is the first report of a small-molecule inhibitor of MEK1/2.

14. Sebolt-Leopold, J. S. et al. Blockade of the MAP

kinase pathway suppresses growth of colon tumors in vivo. Nature Med. 5, 810–816 (1999).

This is the first paper to demonstrate that a potent and specific allosteric MEK1/2 inhibitor could inhibit the growth of tumour xenografts in vivo.15. Frémin, C. & Meloche, S. From basic research to

clinical development of MEK1/2 inhibitors for cancer therapy. J. Hematol. Oncol. 3, 8 (2010).16. Gilmartin, A. G. et al. GSK1120212 (JTP-74057) is an

inhibitor of MEK activity and activation with favorable pharmacokinetic properties for sustained in vivo pathway inhibition. Clin. Cancer Res. 17, 9–1000 (2011).

This study provides a detailed characterization of trametinib, the first MEK1/2 inhibitor to receive US Food and Drug Administration (FDA) approval.17. Davies, H. et al. Mutations of the BRAF gene in human

cancer. Nature 417, 949–9 (2002).

This paper provides the first description of activating mutations in BRAF in human cancer, including the particularly high incidence in melanoma.18. Bollag, G. et al. Clinical efficacy of a RAF inhibitor

needs broad target blockade in BRAF-mutant melanoma. Nature 467, 596–599 (2010).This is the first study that attempted to

systematically quantify the extent of RAF–MEK–ERK suppression required for effective tumour suppression.

19. Bollag, G. Vemurafenib: the first drug approved for

BRAF-mutant cancer. Nat. Rev. Drug Discov. 11, 873–886 (2012).

20. Ascierto, P. A. et al. Phase II trial (BREAK-2) of the

BRAF inhibitor dabrafenib (GSK2118436) in patients with metastatic melanoma. J. Clin. Oncol. 31, 3205–3211 (2013).

21. Holderfield, M., Deuker, M. M., McCormick, F. &

McMahon, M. Targeting RAF kinases for cancer

therapy: BRAF-mutated melanoma and beyond. Nat. Rev. Cancer. 14, 455–467 (2014).

22. Good, M. C., Zalatan, J. G. & Lim, W. A. Scaffold

proteins: hubs for controlling the flow of cellular information. Science 332, 680–686 (2011).

23. Ramos, J. W. The regulation of extracellular signal-regulated kinase (ERK) in mammalian cells. Int. J. Biochem. Cell Biol. 40, 2707–2719 (2008).24. Caunt, C. J. & Keyse, S. M. Dual-specificity MAP

kinase phosphatases (MKPs): shaping the outcome of MAP kinase signalling. FEBS J. 280, 4–504 (2013).

25. Zhang, Z. et al. Dual specificity phosphatase 6

(DUSP6) is an ETS-regulated negative feedback mediator of oncogenic ERK signaling in lung cancer cells. Carcinogenesis 31, 577–586 (2010).26. Rushworth, L. K. et al. Dual-specificity phosphatase 5

regulates nuclear ERK activity and suppresses skin cancer by inhibiting mutant Harvey-Ras (HRasQ61L)-driven SerpinB2 expression. Proc. Natl Acad. Sci. USA 111, 18267–18272 (2014).27. Moriceau, G. et al. Tunable-combinatorial mechanisms

of acquired resistance limit the efficacy of BRAF/MEK cotargeting but result in melanoma drug addiction. Cancer Cell 27, 240–256 (2015).28. Duncan, J. S. et al. Dynamic reprogramming of the

kinome in response to targeted MEK inhibition in triple-negative breast cancer. Cell 149, 307–321 (2012).

The elegant quantitative proteomics approach used in this study reveals the extent of adaptive reprogramming induced by MEK inhibition.29. Fawdar, S. et al. Targeted genetic dependency screen

facilitates identification of actionable mutations in

FGFR4, MAP3K9, and PAK5 in lung cancer. Proc. Natl Acad. Sci. USA 110, 12426–12431 (2013).30. Johannessen, C. M. et al. COT drives resistance to

RAF inhibition through MAP kinase pathway reactivation. Nature 468, 968–972 (2010).

This is the first demonstration of an alternative, non-RAF ‘MEK1/2 activator’ driving BRAFi resistance.

31. Marusiak, A. A. et al. Mixed lineage kinases activate

MEK independently of RAF to mediate resistance to RAF inhibitors. Nat. Commun. 5, 3901 (2014).32. Chiariello, M., Marinissen, M. J. & Gutkind, J. S.

Multiple mitogen-activated protein kinase signaling pathways connect the Cot oncoprotein to the c-jun promoter and to cellular transformation. Mol. Cell. Biol. 20, 1747–1758 (2000).33. Todd, D. E. et al. ERK1/2 and p38 cooperate to induce

a p21CIP1-dependent G1 cell cycle arrest. Oncogene 23, 3284–3295 (2004).

34. Johnson, G. L., Dohlman, H. G. & Graves, L. M. MAPK

kinase kinases (MKKKs) as a target class for small-molecule inhibition to modulate signaling networks and gene expression. Curr. Opin. Chem. Biol. 9, 325–331 (2005).

35. Cuevas, B. D., Abell, A. N. & Johnson, G. L. Role of

mitogen-activated protein kinase kinase kinases in signal integration. Oncogene 26, 3159–3171 (2007).36. Roskoski, R. MEK1/2 dual-specificity protein kinases:

structure and regulation. Biochem. Biophys. Res. Commun. 417, 5–10 (2012).

37. Procaccia, S. & Seger, R. in Encyclopedia of Signaling

Molecules (ed. Choi, S.) 1051–1058 (Springer, 2012).38. Zheng, C. F. & Guan, K. L. Activation of MEK family

kinases requires phosphorylation of two conserved Ser/Thr residues. EMBO J. 13, 1123–1131 (1994).39. Alessi, D. R. Identification of the sites in MAP kinase

kinase-1 phosphorylated by p74raf-1. EMBO J. 13, 1610–1619 (1994).

40. Eblen, S. T., Slack, J. K., Weber, M. J. & Catling, A. D.

Rac-PAK signaling stimulates extracellular signal-regulated kinase (ERK) activation by regulating

formation of MEK1–ERK complexes. Mol. Cell. Biol. 22, 6023–6033 (2002).41. Eblen, S. T. et al. Mitogen-activated protein kinase

feedback phosphorylation regulates MEK1 complex formation and activation during cellular adhesion. Mol. Cell. Biol. 24, 2308–2317 (2004).42. Catalanotti, F. et al. A Mek1–Mek2 heterodimer

determines the strength and duration of the Erk signal. Nat. Struct. Mol. Biol. 16, 294–303 (2009).43. Bélanger, L. F. et al. Mek2 is dispensable for mouse

growth and development. Mol. Cell. Biol. 23, 4778–4787 (2003).44. Giroux, S. et al. Embryonic death of Mek1-deficient

mice reveals a role for this kinase in angiogenesis in the labyrinthine region of the placenta. Curr. Biol. 9, 369–372 (1999).

45. Cowley, S., Paterson, H., Kemp, P. & Marshall, C. J.

Activation of MAP kinase kinase is necessary and sufficient for PC12 differentiation and for

transformation of NIH 3T3 cells. Cell 77, 841–852 (1994).

46. Mansour, S. J. et al. Transformation of mammalian

cells by constitutively active MAP kinase kinase. Science 265, 966–970 (1994).

This paper, along with reference 45, demonstrated that phosphomimetic mutations in the activation loop caused constitutive activation of MEK1; the ‘activated’ MEK1 served as an oncogene when expressed in cells.

47. Solit, D. B. et al. BRAF mutation predicts

sensitivity to MEK inhibition. Nature 439, 358–362 (2006).

This paper studied a panel of human cancer cell lines and showed that BRAF-mutated cells tended to be far more sensitive to MEK inhibition

(addicted to MEK1/2) than those harbouring RAS mutations.

48. Dry, J. R. Transcriptional pathway signatures predict

MEK addiction and response to selumetinib

(AZD6244). Cancer Res. 70, 22–2273 (2010).49. Joseph, E. W. et al. The RAF inhibitor PLX4032

inhibits ERK signaling and tumor cell proliferation in a V600E BRAF-selective manner. Proc. Natl Acad. Sci. USA 107, 14903–14908 (2010).50. Heidorn, S. J. et al. Kinase-dead BRAF and oncogenic

RAS cooperate to drive tumor progression through CRAF. Cell 140, 209–221 (2010).

51. Poulikakos, P. I., Zhang, C., Bollag, G., Shokat, K. M. &

Rosen, N. RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF. Nature 4, 427–430 (2010).52. Hatzivassiliou, G. et al. RAF inhibitors prime wild-type

RAF to activate the MAPK pathway and enhance growth. Nature 4, 431–435 (2010).53. Holderfield, M. et al. RAF inhibitors activate the

MAPK pathway by relieving inhibitory

autophosphorylation. Cancer Cell 23, 594–602 (2013).

The studies in references 50–53 identified the biochemical bases for paradoxical RAF activation by BRAFi, highlighting the need for MEK inhibition for more stable pathway inhibition.. Rodriguez-Viciana, P. et al. Germline mutations in

genes within the MAPK pathway cause cardio-facio-cutaneous syndrome. Science 311, 1287–1290 (2006).

55. Bansal, A., Ramirez, R. D. & Minna, J. D. Mutation

analysis of the coding sequences of MEK-1 and MEK-2 genes in human lung cancer cell lines. Oncogene 14, 1231–1234 (1997).

This is the first report of mutations in MEK1 or MEK2 in human cancer.

56. Estep, A. L., Palmer, C. McCormick, F. & Rauen, K. A.

Mutation analysis of BRAF, MEK1 and MEK2 in 15 ovarian cancer cell lines: implications for therapy. PLoS ONE 2, e1279 (2007).

This is the first report of activating MEK1 or MEK2 mutations in human cancer.57. Marks, J. L. et al. Novel MEK1 mutation identified by

mutational analysis of epidermal growth factor receptor signaling pathway genes in lung

adenocarcinoma. Cancer Res. 68, 5524–5528 (2008).

58. Murugan, A. K., Dong, J., Xie, J. & Xing, M. MEK1

mutations, but not ERK2 mutations, occur in melanomas and colon carcinomas, but none in thyroid carcinomas. Cell Cycle 8, 2122–2124 (2009).

59. Nikolaev, S. I. et al. Exome sequencing identifies

recurrent somatic MAP2K1 and MAP2K2

mutations in melanoma. Nat. Genet. 44, 133–139 (2011).

60. Arcila, M. E. et al. MAP2K1 (MEK1) mutations define

a distinct subset of lung adenocarcinoma associated with smoking. Clin. Cancer Res. 21, 1935–1943 (2015).

61. Blasco, R. B. et al. c-Raf, but not B-Raf, is essential

for development of K-Ras oncogene-driven non-small cell lung carcinoma. Cancer Cell. 19, 652–663 (2011).

62. Karreth, F. A., Frese, K. K., DeNicola, G. M.,

Baccarini, M. & Tuveson, D. A. C-Raf is required for the initiation of lung cancer by K-RasG12D. Cancer Discov. 1, 128–136 (2011).

63. Alessi, D. R., Cuenda, A., Cohen, P., Dudley, D. T. &

Saltiel, A. R. PD 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J. Biol. Chem. 270, 274–27494 (1995).. Favata, M. F. et al. Identification of a novel inhibitor of

mitogen-activated protein kinase kinase. J. Biol. Chem. 273, 18623–18632 (1998).

65. Mody, N., Leitch, J., Armstrong, C., Dixon, J. &

Cohen, P. Effects of MAP kinase cascade inhibitors on the MKK5/ERK5 pathway. FEBS Lett. 502, 21–24 (2001).

66. Squires, M. S., Nixon, P. M. & Cook, S. J. Cell-cycle

arrest by PD184352 requires inhibition of

extracellular signal-regulated kinases (ERK) 1/2 but not ERK5/BMK1. Biochem. J. 366, 673–680 (2002).

NATURE REVIEWS | CANCER

VOLUME 15 | OCTOBER 2015 | 5REVIEWS

67. Ohren, J. F. et al. Structures of human MAP kinase

kinase 1 (MEK1) and MEK2 describe novel

noncompetitive kinase inhibition. Nat. Struct. Mol. Biol. 11, 1192–1197 (2004).

The structural studies of MEK1 and MEK2 in this paper reveal the presence of a unique allosteric inhibitor binding pocket adjacent to, but distinct from, the ATP-binding site; this pocket is the binding site for the allosteric MEKis.68. Yeh, T. C. et al. Biological characterization of

ARRY-142886 (AZD6244), a potent, highly selective mitogen-activated protein kinase kinase 1/2 inhibitor. Clin. Cancer Res. 13, 1576–1583 (2007).69. Fischmann, T. O. et al. Crystal structures of MEK1

binary and ternary complexes with nucleotides and inhibitors. Biochemistry 48, 2661–2674 (2009).70. Rinehart, J. et al. Multicenter phase II study of the

oral MEK inhibitor, CI-1040, in patients with advanced non-small-cell lung, breast, colon, and pancreatic cancer. J. Clin. Oncol. 22, 4456–4462 (2004).71. Smith, C. K. & Windsor, W. T. Thermodynamics of

nucleotide and non-ATP-competitive inhibitor binding to MEK1 by circular dichroism and isothermal titration calorimetry. Biochemistry 46, 1358–1367 (2007).72. Delaney, A. M., Printen, J. A., Chen, H., Fauman, E. B.

& Dudley, D. T. Identification of a novel mitogen-activated protein kinase kinase activation domain recognized by the inhibitor PD 184352. Mol. Cell. Biol. 22, 7593–7602 (2002).73. Friday, B. B. et al. BRAF V600E disrupts

AZD6244-induced abrogation of negative feedback pathways between extracellular signal-regulated

kinase and Raf proteins. Cancer Res. 68, 6145–6153 (2008).

This is the first publication showing that MEKis cause potent MEK1/2 phosphorylation owing to ‘feedback relief’.74. Hatzivassiliou, G. et al. Mechanism of MEK inhibition

determines efficacy in mutant KRAS- versus BRAF-driven cancers. Nature 501, 232–236 (2013).75. Lito, P. et al. Disruption of CRAF-mediated MEK

activation is required for effective MEK inhibition in KRAS mutant tumors. Cancer Cell 25, 697–710 (2014).

This paper and reference 74 demonstrate that newer ‘feedback buster’ MEKis prevent the

MEK1/2 phosphorylation and associated ERK1/2 rebound arising from ‘feedback relief’, providing more durable suppression of KRAS-mutant tumour growth.

76. Pratilas, C. A. et al. V600EBRAF is associated with

disabled feedback inhibition of RAF-MEK signaling and elevated transcriptional output of the pathway. Proc. Natl Acad. Sci. USA 106, 4519–4524 (2009).77. Lito, P. et al. Relief of profound feedback inhibition of

mitogenic signaling by RAF inhibitors attenuates their activity in BRAFV600E melanomas. Cancer Cell 22, 668–682 (2012).78. Ishii, N. et al. Enhanced inhibition of ERK signaling by

a novel allosteric MEK inhibitor, CH5126766, that suppresses feedback reactivation of RAF activity. Cancer Res. 73, 4050–4060 (2013).79. Isshiki, Y. et al. Design and synthesis of novel allosteric

MEK inhibitor CH4987655 as an orally available anticancer agent. Bioorg. Med. Chem. Lett. 21, 1795–1801 (2011).

80. Dickson, M. A. Molecular pathways: CDK4 inhibitors

for cancer therapy. Clin. Cancer Res. 20, 3379–3383 (2014).

81. Smalley, K. S. et al. Increased cyclin D1 expression can

mediate BRAF inhibitor resistance in BRAF

V600E-mutated melanomas. Mol. Cancer Ther. 7, 2876–2883 (2008).82. Aguirre, A. J. et al. Activated Kras and Ink4a/Arf

deficiency cooperate to produce metastatic pancreatic ductal adenocarcinoma. Genes Dev. 17, 3112–3126 (2003).

83. Carragher, L. A. et al. V600EBraf induces gastrointestinal

crypt senescence and promotes tumour progression through enhanced CpG methylation of p16INK4a. EMBO Mol. Med. 2, 458–471 (2010).84. Robinson, J. P. et al. Activated MEK cooperates with

Ink4a/Arf loss or Akt activation to induce gliomas in vivo. Oncogene 30, 1341–1350 (2011).

85. Franco, J., Witkiewicz, A. K. & Knudsen, E. S. CDK4/6

inhibitors have potent activity in combination with pathway selective therapeutic agents in models of pancreatic cancer. Oncotarget 5, 6512–6525 (2014).86. Kwong, L. N. et al. Oncogenic NRAS signaling

differentially regulates survival and proliferation in melanoma. Nat. Med. 18, 1503–1510 (2012).

87. Balmanno, K., Chell, S. D., Gillings, A. S., Hayat, S. &

Cook, S. J. Intrinsic resistance to the MEK1/2 inhibitor AZD6244 (ARRY-142886) is associated with weak ERK1/2 signalling and/or strong PI3K signalling in colorectal cancer cell lines. Int. J. Cancer. 125, 2332–2341 (2009).88. Halilovic, E. et al. PIK3CA mutation uncouples tumor

growth and cyclin D1 regulation from MEK/ERK and mutant KRAS signaling. Cancer Res. 70, 6804–6814 (2010).. Holt, S. V. et al. Enhanced apoptosis and tumor

growth suppression elicited by combination of MEK (selumetinib) and mTOR kinase inhibitors (AZD8055). Cancer Res. 72, 1804–1813 (2012).90. Roberts, P. J. et al. Combined PI3K/mTOR and MEK

inhibition provides broad antitumor activity in faithful murine cancer models. Clin. Cancer Res. 18, 5290–5303 (2012).91. Tolcher, A. W. et al. Antitumor activity in RAS-driven

tumors by blocking AKT and MEK. Clin. Cancer Res. 21, 739–748 (2015).92. Dai, B. et al. STAT3 mediates resistance to MEK

inhibitor through microRNA miR-17. Cancer Res. 71, 3658–3668 (2011).93. Bid, H. K. et al. Development, characterization, and

reversal of acquired resistance to the MEK1 inhibitor selumetinib (AZD6244) in an in vivo model of childhood astrocytoma. Clin. Cancer Res. 19, 6716–6729 (2013).94. Lin, L. et al. The Hippo effector YAP promotes

resistance to RAF- and MEK-targeted cancer therapies. Nat. Genet. 47, 250–256 (2015).95. Flaherty, K. T. et al. Improved survival with MEK

inhibition in BRAF-mutated melanoma. N. Engl. J. Med. 367, 107–114 (2012).96. Turke, A. B. et al. MEK inhibition leads to PI3K/AKT

activation by relieving a negative feedback on ERBB receptors. Cancer Res. 72, 3228–3237 (2012).

97. Mirzoeva, O. K. et al. Subtype-specific MEK-PI3kinase

feedback as a therapeutic target in pancreatic

adenocarcinoma. Mol. Cancer Ther. 12, 2213–2225 (2013).

98. Montero-Conde, C. et al. Relief of feedback inhibition

of HER3 transcription by RAF and MEK inhibitors attenuates their antitumor effects in BRAF-mutant thyroid carcinomas. Cancer Discov. 3, 520–533 (2013).99. Lee, H. J. et al. Drug resistance via feedback activation

of Stat3 in oncogene-addicted cancer cells. Cancer Cell 26, 207–221 (2014).

100. Fiskus, W. et al. BET protein antagonist JQ1 is

synergistically lethal with FLT3 tyrosine kinase

inhibitor (TKI) and overcomes resistance to FLT3-TKI in AML cells expressing FLT-ITD. Mol. Cancer Ther. 13, 2315–2327 (2014).

101. Little, A. S., Smith, P. D. & Cook, S. J. Mechanisms of

acquired resistance to ERK1/2 pathway inhibitors. Oncogene 32, 1207–1215 (2013).

102. Emery, C. M. et al. MEK1 mutations confer resistance

to MEK and B-RAF inhibition. Proc. Natl Acad. Sci. USA 106, 20411–20416 (2009).

This is the first report of acquired resistance to allosteric MEKis; in this case an emergent MEK1 mutation in a patient treated with selumetinib.

103. Hatzivassiliou, G. et al. ERK inhibition overcomes

acquired resistance to MEK inhibitors. Mol. Cancer Ther. 11, 1143–11 (2012).

104. Morris, E. J. et al. Discovery of a novel ERK inhibitor

with activity in models of acquired resistance to BRAF and MEK inhibitors. Cancer Discov. 3, 742–750 (2013).

105. Villanueva, J. et al. Concurrent MEK2 mutation and

BRAF amplification confer resistance to BRAF and

MEK inhibitors in melanoma. Cell Rep. 4, 1090–1099 (2013).

106. Corcoran, R. B. et al. BRAF gene amplification can

promote acquired resistance to MEK inhibitors in

cancer cells harboring the BRAF V600E mutation. Sci. Signal 3, ra84 (2010).

107. Little, A. S. et al. Amplification of the driving

oncogene, KRAS or BRAF, underpins acquired

resistance to MEK1/2 inhibitors in colorectal cancer cells. Sci. Signal 4, ra17 (2011).

This report and reference 106 demonstrate that amplification of BRAFV600E can drive acquired resistance to MEKis, which can be overcome by dual RAF and MEK inhibition. This paper

additionally demonstrates that KRAS amplification can also drive MEKi resistance.

108. Flaherty, K. T. et al. Combined BRAF and MEK

inhibition in melanoma with BRAF V600 mutations. N. Engl. J. Med. 367, 1694–1703 (2012).This is a key clinical proof-of-principle study

demonstrating the potential of dual RAF and MEK inhibition to cause more durable tumour suppression.

109. Infante, J. R. et al. Safety, pharmacokinetic,

pharmacodynamic, and efficacy data for the oral MEK inhibitor trametinib: a phase 1 dose-escalation trial. Lancet Oncol. 13, 773–781 (2012).

110. Musib, L. et al. Clinical pharmacokinetics of

GDC-0973, an oral MEK inhibitor, in cancer patients: data from a Phase 1 study. Cancer Res. 71, 1304 (2011).111. Ribas, A. et al. Combination of vemurafenib and

cobimetinib in patients with advanced BRAFV600-mutated melanoma: a Phase 1b study. Lancet Oncol. 15, 9–965 (2014).

112. Adjei, A. A. et al. Phase I pharmacokinetic and

pharmacodynamic study of the oral, small-molecule mitogen-activated protein kinase kinase 1/2 inhibitor AZD6244 (ARRY-142886) in patients with advanced cancers. J. Clin. Oncol. 26, 2139–2146 (2008).113. Houédé, N. et al. 600 pharmacokinetics and

pharmacodynamics of a selective oral MEK1/2

inhibitor, pimasertib (MSC1936369B/AS703026), in patients with advanced solid tumors. Eur. J. Cancer 48, S184 (2012).

114. Weekes, C. D. et al. Multicenter Phase I trial of the

mitogen-activated protein kinase 1/2 inhibitor BAY 86–9766 in patients with advanced cancer. Clin. Cancer Res. 19, 1232–1243 (2013).

115. Kim, K. B. et al. Phase II study of the MEK1/MEK2

inhibitor Trametinib in patients with metastatic BRAF-mutant cutaneous melanoma previously treated with or without a BRAF inhibitor. J. Clin. Oncol. 31, 482–4 (2013).

116. Long, G. V. et al. Combined BRAF and MEK inhibition

versus BRAF inhibition alone in melanoma. N. Engl. J. Med. 371, 1877–1888 (2014).

117. Robert, C. et al. Improved overall survival in

melanoma with combined dabrafenib and trametinib. N. Engl. J. Med. 372, 30–39 (2015).118. Zimmer, L. et al. Phase I expansion and

pharmacodynamic study of the oral MEK inhibitor RO4987655 (CH4987655) in selected patients with advanced cancer with RAS–RAF mutations. Clin. Cancer Res. 20, 4251–4261 (2014).

119. Bennouna, J. et al. A Phase II, open-label, randomised

study to assess the efficacy and safety of the MEK1/2 inhibitor AZD6244 (ARRY-142886) versus capecitabine monotherapy in patients with colorectal cancer who

have failed one or two prior chemotherapeutic regimens. Invest. New Drugs 29, 1021–1028 (2011).

120. Ascierto, P. A. et al. MEK162 for patients with

advanced melanoma harbouring NRAS or Val600 BRAF mutations: a non-randomised, open-label Phase 2 study. Lancet Oncol. 14, 249–256 (2013).121. Farley, J. et al. Selumetinib in women with recurrent

low-grade serous carcinoma of the ovary or

peritoneum: an open-label, single-arm, Phase 2 study. Lancet Oncol. 14, 134–140 (2013).

122. Bodocky, G. et al. A Phase II open-label randomized

study to assess the efficacy and safety of selumetinib (AZD6244 [ARRY-142886]) versus capecitabine in patients with advanced or metastatic pancreatic

cancer who have failed first-line gemcitabine therapy. Invest. New Drugs 30, 1216–1223 (2012).

123. Bekaii-Saab, T. et al. Multi-institutional Phase II study

of selumetinib in patients with metastatic biliary cancers. J. Clin. Oncol. 29, 2357–2363 (2011).124. Hainsworth, J. D. et al. A phase II, open-label,

randomized study to assess the efficacy and safety of AZD6244 (ARRY-142886) versus pemetrexed in patients with non-small cell lung cancer who have failed one or two prior chemotherapeutic regimens. J. Thorac Oncol. 5, 1630–1636 (2010).

125. Carvajal, R. D. et al. Effect of selumetinib versus

chemotherapy on progression-free survival in uveal melanoma: a randomized clinical trial. JAMA 311, 2397–2405 (2014).

126. Borthakur, G. et al. Phase I/II trial of the MEK1/2

inhibitor GSK1120212 (GSK212) in patients (pts) with relapsed/refractory myeloid malignancies: evidence of activity in pts with RAS mutation. J. Clin. Oncol. 29, 4251–4261 (2011).

127. Jain, N. et al. Phase II study of the oral MEK inhibitor

selumetinib in advanced acute myelogenous leukemia: a University of Chicago Phase II consortium trial. Clin. Cancer Res. 20, 490–498 (2014).

www.nature.com/reviews/cancerREVIEWS

128. Blumenschein, G. Jr et al. A randomized Phase 2

study of the MEK1/MEK2 inhibitor trametinib (GSK1120212) compared with docetaxel in KRAS-mutant advanced non-small cell lung cancer (NSCLC). Ann Oncol. 26, 4–901 (2015).

129. Widemann, B. C. et al. Phase I study of the MEK1/2

inhibitor selumetinib (AZD6244) hydrogen sulfate in children and young adults with neurofibromatosis type 1 (NF1) and inoperable plexiform neurofibromas (PNs). J. Clin. Oncol. 32, S10018 (2014).

130. Banerjee, A. et al. A phase 1 study of AZD6244 in

children with recurrent or refractory low-grade

gliomas: A Pediatric Brain Tumor Consortium report. J. Clin. Oncol. 32, 10065 (2014).

131. Jones, D. T. W. et al. Recurrent somatic alterations of

FGFR1 and NTRK2 in pilocytic astrocytoma Nat. Genet. 45, 927–932 (2013).

132. Corcoran, R. B. et al. Phase 1–2 trial of the BRAF

inhibitor dabrafenib (D) plus MEK inhibitor trametinib (T) in BRAF V600 mutant colorectal cancer (CRC): updated efficacy and biomarker analysis. J. Clin. Oncol. 32, S3517 (2014).

133. Bendell, J. C. et al. Efficacy and tolerability in an open-label phase I/II study of MEK inhibitor trametinib (T), BRAF inhibitor dabrafenib (D), and anti-EGFR

antibody panitumumab (P) in combination in patients (pts) with BRAF V600E mutated colorectal cancer (CRC). J. Clin. Oncol. 32, S3515 (2014).

134. Chen, Z. et al. A murine lung cancer co-clinical trial

identifies genetic modifiers of therapeutic response. Nature 483, 613–617 (2012).

135. Holt, S. V. et al. The MEK1/2 inhibitor, selumetinib

(AZD6244; ARRY-142886), enhances anti-tumour efficacy when combined with conventional chemotherapeutic agents in human tumour xenograft models. Br. J. Cancer 106, 858–866 (2012).

136. Jänne, P. A. et al. Selumetinib plus docetaxel for

KRAS-mutant advanced non-small-cell lung cancer: a randomised, multicentre, placebo-controlled, Phase 2 study. Lancet Oncol. 14, 38–47 (2013).

137. Gandara, D. R. et al. Oral MEK1/MEK2 inhibitor

trametinib (GSK1120212) in combination with docetaxel in KRAS-mutant and wild-type (WT) advanced non-small cell lung cancer (NSCLC): a Phase I/Ib trial. J. Clin, Oncol. 31, S8028 (2013).138. Infante, J. R. et al. A randomised, double-blind,

placebo-controlled trial of trametinib, an oral MEK inhibitor, in combination with gemcitabine for

patients with untreated metastatic adenocarcinoma of the pancreas. Eur. J. Cancer. 50, 2072–2081 (2014).

139. Van Cutsem, E. et al. Phase II randomized trial of

MEK inhibitor pimasertib or placebo combined with gemcitabine in the first-line treatment of metastatic pancreatic cancer. J. Clin. Oncol. 33, S344 (2015).

140. Hochster, H. S. et al. Phase II study of selumetinib

(AZD6244, ARRY-142886) plus irinotecan as second-line therapy in patients with K-RAS mutated colorectal cancer. Cancer Chemother. Pharmacol. 75, 17–23 (2015).

141. Bedard, P. L. et al. A Phase Ib dose-escalation study of

the oral pan-PI3K inhibitor buparlisib (BKM120) in combination with the oral MEK1/2 inhibitor

trametinib (GSK1120212) in patients with selected advanced solid tumors. Clin. Cancer Res. 21, 730–738 (2015).

142. Juric, D. et al. A phase 1b dose-escalation study of

BYL719 plus binimetinib (MEK162) in patients with selected advanced solid tumors. J. Clin. Oncol. 32, S9051 (2014).

143. Chakravarty, D. et al. Small-molecule MAPK inhibitors

restore radioiodine incorporation in mouse thyroid cancers with conditional BRAF activation. J. Clin. Invest. 121, 4700–4711 (2011).

144. Ho, A. L. et al. Selumetinib-enhanced radioiodine

uptake in advanced thyroid cancer. N. Engl. J. Med. 368, 623–632 (2013).

145. Falchook, G. S. et al. Activity of the oral MEK inhibitor

trametinib in patients with advanced melanoma: a Phase 1 dose-escalation trial. Lancet Oncol. 13, 782–7 (2012).

146. Hauschild, A. et al. Dabrafenib in BRAF-mutated

metastatic melanoma: a multicentre, open-label, Phase 3 randomised controlled trial. Lancet 380, 358–365 (2012).

147. Chapman, P. B. et al. Improved survival with

vemurafenib in melanoma with BRAF V600E mutation. N. Engl. J. Med. 3, 2507–2516 (2011).

148. Zhao, Y. & Adjei, A. A. The clinical development of MEK

inhibitors. Nat. Rev. Clin. Oncol. 11, 385–400 (2014).149. Duncan, K. E., Chang, L. Y. & Patronas, M. MEK

inhibitors: a new class of chemotherapeutic agents with ocular toxicity. Eye 29, 1003–1012 (2015).150. Robert, C. et al. Selumetinib plus dacarbazine versus

placebo plus dacarbazine as first-line treatment for BRAF-mutant metastatic melanoma: a phase 2 double-blind randomised study. Lancet Oncol. 14, 733–740 (2013).

151. LoRusso, P. et al. A first-in-human Phase Ib study to

evaluate the MEK inhibitor GDC-0973, combined with the pan-PI3K inhibitor GDC-0941, in patients with advanced solid tumors. J. Clin. Oncol. 30, S2566 (2012).

152. Kurzrock, R. et al. Phase I dose-escalation of the

oral MEK1/2 inhibitor GSK1120212 (GSK212) dosed in combination with the oral AKT inhibitor GSK2141795 (GSK795). J. Clin. Oncol. 29, S3085 (2011).

153. Becerra, C. et al. A five-arm, open-label, Phase I/lb

study to assess safety and tolerability of the oral MEK1/MEK2 inhibitor trametinib (GSK1120212) in combination with chemotherapy or erlotinib in

patients with advanced solid tumors. J. Clin. Oncol. 30, S3023 (2012).

1. Pratilas, C. A. et al. Genetic predictors of MEK

dependence in non-small cell lung cancer. Cancer Res. 68, 9375–9383 (2008).

155. Jing, J. et al. Comprehensive predictive biomarker

analysis for MEK inhibitor GSK1120212. Mol. Cancer Ther. 11, 720–729 (2012).

156. Kelly, K. et al. Oral MEK1/MEK2 inhibitor trametinib

(GSK1120212) in combination with pemetrexed for KRAS-mutant and wild-type (WT) advanced non-small cell lung cancer (NSCLC): a Phase I/Ib trial. J. Clin. Oncol. 31, S8027 (2013).

157. Abdel-Wahab, O. et al. Efficacy of intermittent

combined RAF and MEK inhibition in a patient with concurrent BRAF- and NRAS-mutant malignancies. Cancer Discov. 4, 538–5 (2014).

This is a case study highlighting the use of

combined BRAFis and MEKis to control both target BRAF-mutant melanoma and prevent paradoxical acceleration of incipient RAS-mutant tumours.

158. Loboda, A. et al. A gene expression signature of RAS

pathway dependence predicts response to PI3K and RAS pathway inhibitors and expands the population of RAS pathway activated tumors. BMC Med. Genom. 3, 26 (2010).

159. Lamba, S. et al. RAF suppression synergizes with MEK

inhibition in KRAS mutant cancer cells. Cell Rep. 8, 1475–1483 (2014).

160. Goetz, E. M. Ghandi, M., Treacy, D. J., Wagle, N. &

Garraway, L. A. ERK mutations confer resistance to mitogen-activated protein kinase pathway inhibitors. Cancer Res. 74, 7079–70 (2014).

161. Xing, F. et al. Concurrent loss of the PTEN and RB1

tumour suppressors attenuates RAF dependence in melanomas harboring V600EBRAF. Oncogene 31, 446–457 (2012).

162. Sosman, J. A. et al. A Phase 1b/2 study of LEE011 in

combination with binimetinib in patients with NRAS-mutant melanoma: early encouraging clinical activity. J. Clin. Oncol. 32, S9009 (2014).

163. Sale, M. J. & Cook, S. J. The BH3 mimetic ABT-263

synergizes with the MEK1/2 inhibitor selumetinib/AZD6244 to promote BIM-dependent tumour cell death and inhibit acquired resistance. Biochem. J. 450, 285–294 (2013).

1. Corcoran, R. B. et al. Synthetic lethal interaction of

combined BCL-XL and MEK inhibition promotes tumor regressions in KRAS mutant cancer models. Cancer Cell 23, 121–128 (2013).

165. Liu, L. et al. The BRAF and MEK inhibitors dabrafenib

and trametinib: effects on immune function and in combination with immunomodulatory antibodies targeting PD1, PD-L1 and CTLA-4. Clin. Cancer Res. 21, 1639–1651 (2015).

166. Sturm, O. E. et al. The mammalian MAPK/ERK

pathway exhibits properties of a negative feedback amplifier. Sci. Signal 3, ra90 (2010).

This study uses simplified mathematical models of negative feedback in the RAF–MEK–ERK cascade coupled with wet-laboratory data to explain

resistance to inhibition in normality and why BRAF mutations render the cascade more susceptible to inhibition.

167. Heinrich, R., Neel, B. G. & Rapoport, T. A.

Mathematical models of protein kinase signal transduction. Mol. Cell 9, 957–970 (2002).

168. Voliotis, M., Perrett, R. M., McWilliams, C.

McArdle, C. A. & Bowsher, C. G. Information transfer by leaky, heterogeneous, protein kinase signaling systems. Proc. Natl Acad. Sci. USA 111, E326–E333 (2014).

169. Ebisuya, M., Kondoh, K. & Nishida, E. The duration,

magnitude and compartmentalization of ERK MAP kinase activity: mechanisms for providing signaling specificity. J. Cell Sci. 118, 2997–3002 (2005).170. Kholodenko, B. N., Hancock, J. F. & Kolch, W.

Signalling ballet in space and time. Nat. Rev. Mol. Cell Biol. 11, 414–426 (2010).

171. Rajakulendran, T., Sahmi, M., Lefrançois, M.,

Sicheri, F. & Therrien, M. A dimerization-dependent mechanism drives RAF catalytic activation. Nature 461, 2–5 (2009).

172. Rushworth, L. K., Hindley, A. D., O’Neill, E. & Kolch, W.

Regulation and role of Raf-1/B-Raf heterodimerization Mol. Cell. Biol. 26, 2262–2272 (2006).

173. Dougherty, M. K. et al. Regulation of Raf-1 by direct

feedback phosphorylation. Mol. Cell. 17, 215–224 (2005).

174. Ritt, D. A., Monson, D. M., Specht, S. I. &

Morrison, D. K. Impact of feedback phosphorylation and Raf heterodimerization on normal and mutant B-Raf signaling. Mol. Cell. Biol. 30, 806–819 (2010).

175. Fritsche-Guenther, R. et al. Strong negative feedback

from Erk to Raf confers robustness to MAPK signalling. Mol. Syst. Biol. 7, 4 (2011).

176. McKay, M. M. & Morrison, D. K. Integrating signals

from RTKs to ERK/MAPK. Oncogene 26, 3113–3121 (2007).

177. Yarden, Y. & Slikowski, M. X. Untangling the ErbB

signalling network. Nat. Revs Mol. Cell. Biol. 2, 127–137 (2001).

178. Chen, J., Fujii, K., Zhang, L., Roberts, T. & Fu, H. Raf-1

promotes cell survival by antagonizing apoptosis signal-regulating kinase 1 through a MEK-ERK

independent mechanism. Proc. Natl Acad. Sci. USA 98, 7783–7788 (2001).

179. Mielgo, A. et al. A MEK-independent role for CRAF in

mitosis and tumor progression. Nat. Med. 17, 11–15 (2011).

180. Lorusso, P. M. et al. Phase I and pharmacodynamic

study of the oral MEK inhibitor CI-1040 in patients with advanced malignancies. J. Clin. Oncol. 23, 5281–5293 (2005).

181. LoRusso, P. M. et al. Phase I pharmacokinetic and

pharmacodynamic study of the oral MAPK/ERK kinase inhibitor PD-0325901 in patients with advanced cancers. Clin. Cancer Res. 16, 1924–1937 (2010).182. Lee, P. A. et al. Preclinical development of ARRY-162,

a potent and selective MEK 1/2 inhibitor. Cancer Res. 70, S2515 (2010).

183. Chen, X. et al. Combined PKC and MEK inhibition in

uveal melanoma with GNAQ and GNA11 mutations. Oncogene 33, 4724–4734 (2014).

184. Finn, R. S. et al. A Phase I study of MEK inhibitor

MEK162 (ARRY-438162) in patients with biliary tract cancer. J. Clin. Oncol. 30, S220 (2012).

185. Carter, L. et al. ARRY-162, a novel MEK inhibitor:

results of a 14-day Phase 1a study in healthy subjects and a 28-day Phase 1b study in rheumatoid arthritis patients. American College of Rheumatology [online], https://acr.confex.com/acr/2008/webprogram/Paper3095.html (2008).

186. Banerji, U. et al. The first-in-human study of the

hydrogen sulfate (Hyd-sulfate) capsule of the MEK1/2 inhibitor AZD6244 (ARRY-142886): a Phase I open-label multicenter trial in patients with advanced cancer. Clin. Cancer Res. 16, 1613–1623 (2010).187. Iverson, C. et al. RDEA119/BAY 869766: a potent,

selective, allosteric inhibitor of MEK1/2 for the treatment of cancer. Cancer Res. 69, 6839–6847 (2009).

188. Schmieder, R. et al. Allosteric MEK1/2 inhibitor

refametinib (BAY 86–9766) in combination with sorafenib exhibits antitumor activity in

preclinical murine and rat models of hepatocellular carcinoma. Neoplasia 15, 1161–1171 (2013).

1. Lee, L. et al. The safety, tolerability, pharmacokinetics,

and pharmacodynamics of single oral doses of

CH4987655 in healthy volunteers: target suppression using a biomarker. Clin. Cancer Res. 15, 7368–7374 (2009).

190. Leijen, S. et al. Phase I dose-escalation study of the

safety, pharmacokinetics, and pharmacodynamics of the MEK inhibitor RO4987655 (CH4987655) in

patients with advanced solid tumors. Clin. Cancer Res. 18, 4794–4805 (2012).

NATURE REVIEWS | CANCER

VOLUME 15 | OCTOBER 2015 | 591REVIEWS

191. Kim, K. et al. Blockade of the MEK/ERK signalling

cascade by AS703026, a novel selective MEK1/2

inhibitor, induces pleiotropic anti-myeloma activity in vitro and in vivo. Br. J. Haematol. 149, 537–9 (2010).192. Dong, Q. et al. Discovery of TAK-733, a potent and

selective MEK allosteric site inhibitor for the treatment of cancer. Bioorg. Med. Chem. Lett. 21, 1315–1319 (2011).

193. Nakamura, A. et al. Antitumor activity of the selective

pan-RAF inhibitor TAK-632 in BRAF inhibitor-resistant melanoma. Cancer Res. 73, 7043–7055 (2013).

194. Adjei, A. A. et al. Multicenter dose-escalation study of

the investigational drug TAK-733, an oral MEK inhibitor in patients with advanced solid tumors; preliminary phase 1 result. Oncology PRO [online], http://oncologypro.esmo.org/Meeting-Resources/ESMO-2012/Multicenter-dose-escalation-study-of-the-investigational-drug-TAK-733-an-oral-MEK-inhibitor-in-patients-pts-with-advanced-solid-tumors-preliminary-phase-1-results (2012).

195. Leonowens, C. et al. Concomitant oral and

intravenous pharmacokinetics of trametinib, a MEK inhibitor, in subjects with solid tumours. Br. J. Clin. Pharmacol. 78, 524–532 (2014).

196. Martinez-Garcia, M. et al. First-in-human, Phase I

dose-escalation study of the safety, pharmacokinetics, and pharmacodynamics of RO5126766, a

first-in-class dual MEK/RAF inhibitor in patients with solid tumors. Clin. Cancer Res. 18, 4806–4819 (2012).

197. Hoeflich, K. P. et al. Intermittent administration of

MEK inhibitor GDC-0973 plus PI3K inhibitor GDC-0941 triggers robust apoptosis and tumor growth inhibition. Cancer Res. 72, 210–219 (2012).

198. El-Khoueiry, A. et al. A first in-human Phase I study to

evaluate the MEK1/2 inhibitor GDC-0623 in patients with advanced solid tumors. Mol. Cancer Ther. 12, B75 (2013).

Acknowledgements

The authors thank J. Sebolt-Leopold who provided informa-tion as a personal communication and apologize to those whose work was not included owing to space limitations.

Competing interests statement

The authors declare competing interests: see Web version for details.

DATABASESNCT01750281: https://clinicaltrials.gov/ct2/show/NCT01750281NCT01933932: https://clinicaltrials.gov/ct2/show/NCT01933932SUPPLEMENTARY INFORMATIONSee online article: S1 (figure)ALL LINKS ARE ACTIVE IN THE ONLINE PDF592 | OCTOBER 2015 | VOLUME 15 © 2015 Macmillan Publishers Limited. All rights reserved

www.nature.com/reviews/cancer

因篇幅问题不能全部显示，请点此查看更多更全内容

查看全文