Resistance to the RAF inhibitor vemurafenib arises commonly in melanomas driven by the activated BRAF oncogene. Here, we report antitumor properties of RAF709, a novel ATP-competitive kinase inhibitor with high potency and selectivity against RAF kinases. RAF709 exhibited a mode of RAF inhibition distinct from RAF monomer inhibitors such as vemurafenib, showing equal activity against both RAF monomers and dimers. As a result, RAF709 inhibited MAPK signaling activity in tumor models harboring either BRAFV600 alterations or mutant N- and KRAS-driven signaling, with minimal paradoxical activation of wild-type RAF. In cell lines and murine xenograft models, RAF709 demonstrated selective antitumor activity in tumor cells harboring BRAF or RAS mutations compared with cells with wild-type BRAF and RAS genes. RAF709 demonstrated a direct pharmacokinetic/pharmacodynamic relationship in in vivo tumor models harboring KRAS mutation. Furthermore, RAF709 elicited regression of primary human tumor–derived xenograft models with BRAF, NRAS, or KRAS mutations with excellent tolerability. Our results support further development of inhibitors like RAF709, which represents a next-generation RAF inhibitor with unique biochemical and cellular properties that enables antitumor activities in RAS-mutant tumors.

Significance: In an effort to develop RAF inhibitors with the appropriate pharmacological properties to treat RAS mutant tumors, RAF709, a compound with potency, selectivity, and in vivo properties, was developed that will allow preclinical therapeutic hypothesis testing, but also provide an excellent probe to further unravel the complexities of RAF kinase signaling. Cancer Res; 78(6); 1537–48. ©2018 AACR.

The MAPK signaling pathway, comprised of H/K/N-RAS, A/B/C-RAF, MEK1/2, and ERK1/2, plays a major role in the regulation of cellular functions such as cell-cycle regulation, proliferation, survival, and migration. The pathway is activated by extracellular signals that in turn induces the small G protein RAS to exchange GDP for GTP, and results in activation of the RAF/MEK/ERK cascade. The MAPK pathway is activated in many human cancers such as those harboring mutations in RAS or RAF. The RAS genes are the most frequently mutated oncogenes in all cancers (reviewed in ref. 1); however, therapeutic targeting of RAS proteins has been challenging. Recent studies described small molecules that specifically target the KRASG12C mutation through covalent inhibition and result in cellular active inhibitors in the low micromolar range, thus offering a potential to develop therapeutics for KRASG12C-driven tumors (2–4). Inhibitors developed that target downstream effectors of RAS, such as RAF, MEK, and ERK kinases, have not demonstrated significant clinical activity in RAS-driven tumors. RAF inhibitors such as vemurafenib and dabrafenib, which are efficacious in melanomas with BRAFV600 mutations, are ineffective in RAS-mutant cancers and instead induce paradoxical activation in BRAF wild-type cells through transactivation of stabilized RAF dimers (5–7). It has been recently demonstrated that these inhibitors have decreased affinity for the second RAF protomer due to negative cooperativity (8). A wide array of inhibitors targeting MEK has also been developed with several under clinical evaluation (reviewed in ref. 9). Although these inhibitors have demonstrated efficacy in preclinical RAS-mutant tumor models, they have not shown clear clinical benefits in patients with tumors harboring RAS mutations as monotherapy, likely due to a narrow therapeutic index and feedback-mediated pathway reactivation (10).

Emerging biology illustrating how RAS proteins activate their downstream targets has provided new opportunities for therapeutic development. CRAF (RAF1) was recently demonstrated as the critical mediator of mutant KRAS-driven cell proliferation and tumor development in both cellular and genetically engineered mouse tumor models (11, 12). CRAF was also shown to be the mediator of feedback-mediated pathway reactivation following MEK inhibitor treatment in KRAS-mutant cancers (10, 13). In addition, CRAF plays an essential role in mediating paradoxical activation following BRAF inhibitor treatment (5–7). Thus selective inhibitors that potently inhibit the activity of CRAF could be both effective in blocking mutant RAS-driven tumorigenesis and alleviating feedback activation. This notion is supported by the recent report of LY3009120, a pan-RAF inhibitor that effectively inhibits active RAF homo-and heterodimers and exhibits activities in tumors with RAS mutations (14). However, in addition to RAF isoforms, LY3009120 potently inhibits several other kinases, including those that have important biological functions such as Ephrin receptors, JNK, p38, and SRC family members. Although the clinical results from LY3009120 have not been published, it is clear from studies with other RAF inhibitors that off-target kinase inhibition can limit dose escalation. For example, RAF265 and BGB-283 inhibit multiple kinases at concentrations similar to the RAFs and their toxicities are suggestive of receptor tyrosine kinase inhibition (e.g., thrombocytopenia, hypertension; refs. 15–17). We hypothesized that a highly selective RAF dimer inhibitor would be valuable to assess on-target antitumor activities of this new class of molecules. In this report, we describe the pharmacologic characterization of RAF709, a type 2 ATP-competitive inhibitor we have developed that potently inhibits RAF kinases with high selectivity. RAF709 demonstrates equal potency in inhibiting RAF monomers and dimers, and exhibits inhibition of MAPK signaling in tumor models harboring BRAF or N/KRAS mutations with minimal paradoxical activation. Correspondingly, RAF709 exhibits greater antitumor activity in cell line and tumor xenograft models harboring BRAF or RAS mutations as compared with those that are wild-type. Furthermore, RAF709 in combination with a MEK inhibitor leads to increased antitumor activity in RAS-mutant models that are insensitive to RAF709 single agent.

RAF in vitro enzyme assays

Enzymatic activities of purified B/CRAF proteins were measured using inactive MEK1 protein as a substrate. Substrate phosphorylation was detected using the anti-pMEK1/2 (S217/S221) antibody, AlphaScreen Protein A coated acceptor beads and streptavidin coated donor beads, and read in an EnVision reader. To measure the inhibitory activity of RAF709, compound was added to the enzyme assay plates with the final concentration from 25 to 1.74E-6 μmol/L. Kinase selectivity of RAF709 was determined using the KINOMEscan screening platform (DiscoverX) that quantitatively measures interactions between RAF709 and 456 human kinases.

Immunoprecipitation

Cells were seeded in 15-cm dishes and incubated with the indicated concentrations of compound for 1 hour. Cell lysates were prepared in immunoprecipitation buffer supplemented with 1× protease and 1× phosphatase inhibitor cocktails. Cleared lysates were normalized for protein concentration and incubated with specific antibody overnight at 4°C. Protein A Ultra Link Resin (Thermo Scientific) was then added to each sample and incubated for 2 hours at 4°C. Resin was washed with immunoprecipitation lysis buffer before bound proteins were eluted in SDS sample buffer.

Cell proliferation

All cell lines were purchased from commercial sources and maintained and as described previously (18). Cell lines were confirmed to be mycoplasma-free by PCR detection and authenticated by SNP genotyping. Cells were seeded in 384-well plates and incubated with a 1:3 serial dilution of compound starting from 30 μmol/L for 5 days. Cell viability was measured using the Cell Titer-Glo Luminescent Cell Viability Assay Kit (Promega) and read on an Envision (Perkin Elmer) plate reader. Percent growth inhibition was calculated by normalizing treatment to DMSO control. IC50 values, where cell growth was inhibited by 50%, were calculated from dose–response curves generated in GraphPad PRISM using a nonlinear regression 4-parameter curve fitting model.

Combination activity

Combination testing was performed using a large-scale compound screening platform as described previously (19). Cells were plated in triplicate into 384-well plates and treated with compounds serially diluted 1:3 across a dose range. Following 5 days of compound treatment, cell viability was determined as described above. Compound combination effects were assessed by weighted “Synergy Scores” calculated using the Loewe dose additivity model. In addition, isobologram and combination index values were generated using the same model to better interpret the doses at which the most profound synergy effects were seen.

In vivo pharmacodynamics and efficacy

Mice were maintained and handled in accordance with the Novartis Institutes for BioMedical Research (NIBR) Institutional Animal Care and Use Committee (IACUC) and all studies were approved by the NIBR IACUC. Calu-6 and HPAFII tumor xenografts were generated by implanting cells in 50% Matrigel subcutaneously into the right flank of female nude mice (6–8 weeks old). For tumor pharmacodynamics measurements, tumor-bearing mice were randomized into treatment groups and treated with indicated inhibitor. Tumor samples were collected at different time points after single dose (n = 3/time point) and analyzed for levels of phospho- and total MEK1/2 using the MesoScale Discovery (MSD) platform or DUSP6 mRNA by qPCR. For the in vivo efficacy study, mice were randomized into treatment groups. Tumor volume and body weights were collected at the time of randomization and twice per week for the study duration. Tumor volume was determined by measurement with calipers and calculated using a modified ellipsoid formula, where tumor volume (TV) (mm3) = [((l × w2) × 3.14159))/6], where l is the longest axis of the tumor and w is perpendicular to l. The general health of mice was monitored daily and behavior and well-being were monitored twice weekly.

Efficacy in primary tumor models

Patient-derived tumor xenograft models (PDX) models were established and characterized as described previously (20). Tumor response was determined by comparing tumor volume change at time measured (t) to its baseline: % tumor volume change = ΔVolt = 100% × ((Vt – Vinitial)/Vinitial). The BestResponse was the minimum value of ΔVolt for t ≥ 10 d. For each time t, the average of ΔVolt from t = 0 to t was also calculated. We defined the BestAvgResponse as the minimum value of this average for t ≥ 10 d. This metric captures a combination of speed, depth, and durability of response into a single value.

Discovery of RAF709 as a highly selective RAF kinase inhibitor

By leveraging the existing knowledge of targeting RAF and structure-based rational design, we have developed RAF709, N-(2-methyl-5′-morpholino-6′-((tetrahydro-2H-pyran-4-yl)oxy)-[3,3′-bipyridin]-5-yl)-3-(trifluoromethyl)benzamide, a highly active and selective inhibitor of RAF isoforms. In in vitro biochemical assays, RAF709 exhibited potent inhibitory activity targeting BRAF, BRAFV600E, and CRAF with IC50 values ranging between 0.3 to 1.5 nmol/L (Fig. 1). Data obtained from the cocrystal structure of RAF709 in complex with the BRAF kinase domain revealed that the protein adopts an inactive conformation with the DFG out and the αC-helix in, characteristic of a type II inhibitor binding mode (21). Kinase selectivity of RAF709 was evaluated using the KINOMEscan screening platform, which quantitatively measured the binding of RAF709 against 456 human kinases and their mutated derivatives (22). Of the 456 kinases tested, RAF709 showed a high degree of selectivity for RAF kinases with only BRAF, BRAFV600E, CRAF, and DDR1 demonstrating greater than 99% binding at 1 μmol/L (Fig. 1; Supplementary Tables S1 and S2). Three additional kinases, PDGFRB, FRK, and DDR2, showed greater than 85% binding by RAF709. The remaining 449 kinases all showed less than 65% binding by RAF709 (Supplementary Tables S1 and S2). While we have not confirmed the potency against these off-target kinases in biochemical or cellular assay, the kinase selectivity of RAF709 was further assessed using the KiNativ platform, in which the cellular selectivity is determined by competition with an ATP-competitive covalent probe and read-out by mass spectrometry (23). In HCT116 cells treated for 2 hours at 10 μmol/L RAF709, the only kinases inhibited ≥ 80% were BRAF and CRAF, the next most potently inhibited kinases were ARAF (47%) and EPHA2 (58%) and of the remaining 253 kinases detected, including FRK, none were inhibited >50% (Supplementary Table S3). The apparent inferior binding to ARAF compared with BRAF and CRAF was difficult to confirm in biochemical kinase assays due to the low enzymatic activity of recombinant purified ARAF; however, the potency was in-line with the potency against BRAF and CRAF. The KiNativ data further confirm the exquisite selectivity of RAF709 and provide a comparative dataset to LY3009120, which was evaluated using the same platform in A375 cell lysates and inhibited 17 kinases >50% at concentrations less than 1 μmol/L (14). The cellular selectivity of RAF709 was also evaluated in a Ba/F3 cell panel using wild-type or Ba/F3 cells rendered IL3 independent by stably expressing 38 different kinase oncogenes (24). Consistent with on-target activity against RAF kinases, RAF709 was active in Ba/F3 cells expressing the BRAFV600E oncogene with an IC50 of 0.52 μmol/L with little activity observed in cells expressing 36 additional kinases (Supplementary Table S4). Collectively, these data demonstrated that RAF709 is an active and highly selective inhibitor targeting the RAF kinases.

Figure 1.

RAF709 exhibits highly selective activity targeting both BRAF and CRAF kinases. Chemical structure of RAF709. Activity of RAF709 against BRAF, CRAF, and BRAFV600E presented as IC50 values measured in in vitro biochemical assays and percentage of target binding at 1 μmol/L of RAF709 in a binding assay as part of the KINOMEscan human kinase panel profiling. RAF709 kinase selectivity profile is represented on the human kinase phylogenetic tree; targets bound are marked with red circles, with the size of the circles proportional to percentage of binding to each kinase.

Figure 1.

RAF709 exhibits highly selective activity targeting both BRAF and CRAF kinases. Chemical structure of RAF709. Activity of RAF709 against BRAF, CRAF, and BRAFV600E presented as IC50 values measured in in vitro biochemical assays and percentage of target binding at 1 μmol/L of RAF709 in a binding assay as part of the KINOMEscan human kinase panel profiling. RAF709 kinase selectivity profile is represented on the human kinase phylogenetic tree; targets bound are marked with red circles, with the size of the circles proportional to percentage of binding to each kinase.

Close modal

RAF709 is an effective inhibitor of BRAF monomers and RAF dimers in BRAFmut and KRASmut tumor cells

The pharmacologic activity of RAF inhibitors in cells expressing different BRAF mutations and mutant RAS have highlighted the role of RAF dimers in signaling and inhibitor sensitivity. BRAFV600 mutants are activated monomers and sensitive to vemurafenib and other RAF inhibitors in this class such as dabrafenib and encorafenib, whereas other activating BRAF mutants function as RAS dependent heterodimers with CRAF (8). These inhibitors are inactive against RAF dimers because they have much reduced affinity for the second protomer when the first protomer is drug-bound. We used the cellular system developed by Yao and colleagues to examine the activity of RAF709 in inhibiting RAF monomers and dimers in tumor cells harboring BRAFV600E (A375) or KRASG13D (HCT116; Fig. 2A; Supplementary Fig. S1). At a concentration of 1 μmol/L, encorafenib induces maximum activation of pMEK/pERK in HCT116 cells and at this concentration we presume that one protomer from each active RAF dimer is occupied, while the second protomer is free to phosphorylate MEK (8). Encorafenib has a very slow-off rate (25) and therefore excess drug can be washed out while maintaining maximal activation through occupancy on the first protomer. Following wash-out, cells were treated with a dose range of either dabrafenib or RAF709 for an additional hour. Phosphorylation of MEK and ERK were determined by Western blot analysis (Supplementary Fig. S1) and quantified to determine the dose response (Fig. 2A). In A375 (BRAFV600E) cells, both dabrafenib and RAF709 showed robust activity inhibiting mutant BRAF monomer–driven ERK activation with IC50s of 5 nmol/L and 44 nmol/L, respectively. In comparison, in HCT116 cells (KRASG13D) pretreated with encorafenib, RAF709 exhibited equipotent activity inhibiting RAF dimer–driven signaling with a pERK IC50 of 79 nmol/L, whereas dabrafenib showed approximately 100-fold higher IC50 of 3 μmol/L. The same experiment was carried out in SK-MEL-30 (NRASQ61K) cells with similar results: RAF709 had an IC50 of 0.139 μmol/L (3.2-fold shift) and dabrafenib had an IC50 of 3.7 μmol/L (>700-fold shift; Supplementary Fig. S2). These data suggest that RAF709 inhibits both RAF monomers and dimers with similar potency.

Figure 2.

RAF709 is an effective inhibitor of BRAF monomers and RAF dimers. A, HCT116 (KRASG13D) cells were treated with DMSO or 1 μmol/L encorafenib for 1 hour. After the first RAF site was occupied by encorafenib, encorafenib was then washed off from the cells and replaced with dabrafenib or RAF709 across a dose range for 1 hour. A375 (BRAFV600E) cells were treated with dabrafenib or RAF709 across a dose range for 1 hour. pERK levels were assessed by Western blot analysis (Supplementary Fig. S1) and quantified to generate the dose inhibition curves and IC50 values in A375 and SK-MEL-30 cells, representing the compound inhibition of BRAFV600E monomer second site and wild-type RAF dimer second site, respectively. Potency of the inhibitor in inhibiting RAF dimer relative to monomer is represented as the ratio of dimer second-site pERK IC50 versus monomer second-site pERK IC50. B, A375 cells expressing the doxycycline (Dox)-inducible p61BRAFV600E were maintained in culture media without or with doxycycline for 2 days, then treated with DMSO, dabrafenib, or RAF709 at the concentrations indicated for 2 hours. Cell lysates were prepared for Western blot analysis of protein levels of BRAFV600E, pMEK, and pERK. GAPDH was included as a loading control. For cell proliferation assays, cells were maintained in culture media without or with doxycycline for 2 days, followed by treatment with dabrafenib or RAF709 for 5 additional days across a dose range to determine IC50. C, HCT116 cells were treated with DMSO or RAF709 at indicated concentrations for 1 hour; 1 μmol/L dabrafenib treatment was included for comparison. BRAF/CRAF dimerization was assessed by immunoprecipitating BRAF or CRAF, followed by Western blot analysis of BRAF and CRAF. Levels of pMEK and pERK in whole-cell lysates (WCL) were determined by Western blot analysis. GAPDH level was included as a loading control.

Figure 2.

RAF709 is an effective inhibitor of BRAF monomers and RAF dimers. A, HCT116 (KRASG13D) cells were treated with DMSO or 1 μmol/L encorafenib for 1 hour. After the first RAF site was occupied by encorafenib, encorafenib was then washed off from the cells and replaced with dabrafenib or RAF709 across a dose range for 1 hour. A375 (BRAFV600E) cells were treated with dabrafenib or RAF709 across a dose range for 1 hour. pERK levels were assessed by Western blot analysis (Supplementary Fig. S1) and quantified to generate the dose inhibition curves and IC50 values in A375 and SK-MEL-30 cells, representing the compound inhibition of BRAFV600E monomer second site and wild-type RAF dimer second site, respectively. Potency of the inhibitor in inhibiting RAF dimer relative to monomer is represented as the ratio of dimer second-site pERK IC50 versus monomer second-site pERK IC50. B, A375 cells expressing the doxycycline (Dox)-inducible p61BRAFV600E were maintained in culture media without or with doxycycline for 2 days, then treated with DMSO, dabrafenib, or RAF709 at the concentrations indicated for 2 hours. Cell lysates were prepared for Western blot analysis of protein levels of BRAFV600E, pMEK, and pERK. GAPDH was included as a loading control. For cell proliferation assays, cells were maintained in culture media without or with doxycycline for 2 days, followed by treatment with dabrafenib or RAF709 for 5 additional days across a dose range to determine IC50. C, HCT116 cells were treated with DMSO or RAF709 at indicated concentrations for 1 hour; 1 μmol/L dabrafenib treatment was included for comparison. BRAF/CRAF dimerization was assessed by immunoprecipitating BRAF or CRAF, followed by Western blot analysis of BRAF and CRAF. Levels of pMEK and pERK in whole-cell lysates (WCL) were determined by Western blot analysis. GAPDH level was included as a loading control.

Close modal

To further test the activity of RAF709 in inhibiting dimerized RAF, we took advantage of the p61BRAFV600E mutant that signals as a constitutive dimer (26) and asked whether RAF709 would remain active in cells expressing the p61BRAFV600E mutant. We generated A375 cells expressing p61BRAFV600E in a doxycycline-inducible system, and measured the activity of RAF709 in inhibiting pathway signaling and proliferation in the absence or presence of doxycycline induction (Fig. 2B). The activity of dabrafenib was significantly reduced by the expression of the p61 V600E constitutive dimer, leading to more than 1,000-fold dose response shift in both signaling and growth inhibition. In contrast, RAF709 activity was minimally affected by p61 V600E with only approximately 5-fold IC50 shift in both pMEK/pERK inhibition and inhibition of proliferation. These results further strengthen the hypothesis that RAF709 is a potent inhibitor of both RAF monomers and dimers.

We then investigated how RAF709 would affect B/CRAF heterodimerization and downstream signaling in the KRASG13D HCT116. B/CRAF dimerization in the cells was assessed by immunoprecipitating one RAF isoform followed by Western blotting with the second isoform. Both co-IP studies showed that RAF709 treatment led to a dose-dependent induction of B/CRAF heterodimerization in HCT116, but inhibited MEK and ERK phosphorylation, in line with the ability of RAF709 to effectively inhibit the RAF dimers (Fig. 2C). In contrast, dabrafenib induced B/CRAF dimerization and increased both MEK and ERK phosphorylation. These data demonstrate that RAF709 exhibits a mode of inhibition distinct from the class of RAF monomer inhibitors. Its activity inhibiting both RAF monomers and dimers suggests it should be effective in treating tumors harboring BRAF or RAS mutations.

RAF709 selectively inhibits oncogenic signaling and proliferation in tumor cells with BRAF, NRAS, or KRAS mutations with minimal paradoxical activation

We next examined the activity of RAF709 and dabrafenib in cell lines harboring BRAFV600, NRAS, or KRAS mutations (Fig. 3A). In A375 cells, both dabrafenib and RAF709 inhibited MEK and ERK phosphorylation to near completion at 0.05 and 0.5 μmol/L, respectively. In contrast, in the three cell lines harboring either NRAS or KRAS mutation, dabrafenib treatment at 0.05 and 0.5 μmol/L led to an increase in MEK and ERK phosphorylation, and only at 5 μmol/L showed modest inhibition. In comparison, RAF709 showed dose-dependent inhibition of MEK and ERK phosphorylation without apparent pathway activation in all three RAS-mutant models (minimal activation of pMEK in IPC-298 and HCT116). The ability of RAF709 to inhibit pathway signaling was comparable in cells harboring different RAS mutations and those with the BRAFV600 mutation, reaching near complete inhibition of pMEK and pERK at 0.5 μmol/L.

Figure 3.

RAF709 inhibits oncogenic signaling and proliferation in tumor cells with BRAF, NRAS, and KRAS mutations, with minimal paradoxical activation. A, Cell lines harboring different BRAF or RAS mutations were treated with DMSO, dabrafenib, or RAF709 at the indicated concentrations for 2 hours. Inhibition of MEK or ERK phosphorylation was measured by Western blot analysis. B, Growth inhibition of cell lines after 5 days of treatment by RAF709 (green curve) or dabrafenib (blue curve) was determined and IC50 values are presented in the table.

Figure 3.

RAF709 inhibits oncogenic signaling and proliferation in tumor cells with BRAF, NRAS, and KRAS mutations, with minimal paradoxical activation. A, Cell lines harboring different BRAF or RAS mutations were treated with DMSO, dabrafenib, or RAF709 at the indicated concentrations for 2 hours. Inhibition of MEK or ERK phosphorylation was measured by Western blot analysis. B, Growth inhibition of cell lines after 5 days of treatment by RAF709 (green curve) or dabrafenib (blue curve) was determined and IC50 values are presented in the table.

Close modal

We next determined the antiproliferation activity of RAF709 and dabrafenib in these cell lines. In line with the signaling data, dabrafenib showed the most potent antiproliferative activity in BRAFV600E A375 cells (IC50 = 0.003 μmol/L), approximately 100-fold less potent in NRASmut IPC-298 cells (IC50 = 0.27 μmol/L), and very little activity in the two KRASmut cell lines (Calu-6 IC50 = 23 μmol/L, HCT116 IC50 = 17 μmol/L; Fig. 3B). In contrast, RAF709 showed less IC50 dose shift in RASmut models compared with the BRAFmut model and exhibited dose-dependent growth inhibition in all cell lines examined, with IC50 values of 0.13 μmol/L, 0.07 μmol/L, 1.4 μmol/L, and 0.98 μmol/L in A375, IPC-298, Calu-6, and HCT116, respectively. We also examined the mechanism by which RAF709 inhibited tumor cell proliferation. Cell-cycle analysis by FACS in both Calu-6 and HCT116 after 48 hours of RAF709 treatment indicated that RAF709 led to cell-cycle arrest in the G1 phase with a concomitant increase of cells in sub-G1, indicating cell death (Supplementary Fig. S3A). In both Calu-6 and HCT116 cells, the maximum G1 induction was observed at 1.1 μmol/L with increasing percent of cell death at ≥3.3 μmol/L. In agreement with the cell-cycle results, Western blot analysis of Calu-6 and HCT116 cells following 48 hours of inhibitor treatment showed inhibition of MEK and ERK phosphorylation, and induction of p27 and cleaved-PARP by RAF709 in a dose-dependent manner (Supplementary Fig. S3B).

To confirm the anticancer activity of RAF709 observed in RASmut cells is mediated by on-target inhibition of the RAF kinase function, we expressed a constitutively active variant of MEK1, MEK DD (MEK1 S217D/S221D) under doxycycline regulation in Calu-6 cells, and examined whether MEK DD expression could rescue RAF709 activity in Calu-6 cells (Fig. 4A and B). Without MEK DD expression (−Dox), RAF709 showed dose-dependent inhibition of MEK and ERK phosphorylation and concomitant increase in cPARP (Fig. 4A). In comparison, RAF709 was unable to suppress constitutively activated MEK signaling (+Dox) up to the highest concentration tested at 10 μmol/L. Corresponding to the signaling data, cell growth in the absence of MEK DD induction was sensitive to RAF709 inhibition but insensitive in the presence of MEK DD expression (Fig. 4B). Overexpression of MEK DD alone had no impact on the growth of Calu-6 cells and these data support that the activity RAF709 exhibited in RASmut cells is mediated by on-target inhibition of the oncogenic RAF/MEK/ERK signaling.

Figure 4.

RAF709 demonstrates selective anticancer activity in KRASmut NSCLC cells. A, Calu-6 cells expressing doxycycline-inducible constitutively active MEK (MEK DD) were treated with or without doxycycline for 2 days, followed by DMSO or RAF709 treatment at indicated concentrations for 24 hours. MEK/ERK phosphorylation and PARP cleavage were determined by Western blot analysis. B, Growth inhibition of Calu-6 cells by RAF709 without (blue curve) or with (red curve) doxycycline (Dox)-inducible expression of MEK DD was measured after 3 days of inhibitor treatment across a dose range. IC50 values are indicated in the table.

Figure 4.

RAF709 demonstrates selective anticancer activity in KRASmut NSCLC cells. A, Calu-6 cells expressing doxycycline-inducible constitutively active MEK (MEK DD) were treated with or without doxycycline for 2 days, followed by DMSO or RAF709 treatment at indicated concentrations for 24 hours. MEK/ERK phosphorylation and PARP cleavage were determined by Western blot analysis. B, Growth inhibition of Calu-6 cells by RAF709 without (blue curve) or with (red curve) doxycycline (Dox)-inducible expression of MEK DD was measured after 3 days of inhibitor treatment across a dose range. IC50 values are indicated in the table.

Close modal

RAF709 exhibits increased potency in cancer cell lines harboring BRAF or RAS mutations

We next profiled the antiproliferative activity of RAF709 in a broad panel of genetically characterized human cancer cell line models (18). Using an IC50 value of 5 μmol/L based on in vivo exposure (Cav) at well-tolerated doses, we determined the number of sensitive (IC50 < 5 μmol/L) and insensitive (IC50 > 5 μmol/L) cell lines (Fig. 5A). Cells harboring BRAF, KRAS, or NRAS mutations exhibited significantly increased sensitivity compared with those that are wild-type, with P values of 2.41 × 10−26, 4.08 × 10−4, and 1.73 × 10−7, respectively (Fisher exact test).

Figure 5.

RAF709 exhibits greater antiproliferative activity in cancer cell lines harboring BRAF or RAS mutations. A, Dot plots of IC50 values for growth inhibition in 352 human cancer cell lines by RAF709 following 3 days of inhibitor treatment. The dotted line represents an IC50 of 5 μmol/L, which was used as a cutoff for cell line sensitivity based on in vivo drug exposure from pharmacology studies (Cav ∼ 5 μmol/L at 100 mg/kg). The number of sensitive and resistant cell lines to each inhibitor among BRAFmut, KRASmut, NRASmut, or wild-type (WT) cells is indicated below the graph. A Fisher exact test was performed to determine the statistical significance of inhibitor activity in BRAF- or RAS-mutant cell lines versus WT cell lines. B, The antiproliferative activity of RAF709 was compared with the selective MEK inhibitor trametinib, selective RAF inhibitor dabrafenib, and the relatively nonselective RAF inhibitors sorafenib and RAF265. As in A, each dot represents a different tumor cell line. Purple dots, BRAF-mutant cells; pink dots, KRAS-mutant cells; green dots, NRAS-mutant cells; and black dots, wild-type RAS and BRAF cells.

Figure 5.

RAF709 exhibits greater antiproliferative activity in cancer cell lines harboring BRAF or RAS mutations. A, Dot plots of IC50 values for growth inhibition in 352 human cancer cell lines by RAF709 following 3 days of inhibitor treatment. The dotted line represents an IC50 of 5 μmol/L, which was used as a cutoff for cell line sensitivity based on in vivo drug exposure from pharmacology studies (Cav ∼ 5 μmol/L at 100 mg/kg). The number of sensitive and resistant cell lines to each inhibitor among BRAFmut, KRASmut, NRASmut, or wild-type (WT) cells is indicated below the graph. A Fisher exact test was performed to determine the statistical significance of inhibitor activity in BRAF- or RAS-mutant cell lines versus WT cell lines. B, The antiproliferative activity of RAF709 was compared with the selective MEK inhibitor trametinib, selective RAF inhibitor dabrafenib, and the relatively nonselective RAF inhibitors sorafenib and RAF265. As in A, each dot represents a different tumor cell line. Purple dots, BRAF-mutant cells; pink dots, KRAS-mutant cells; green dots, NRAS-mutant cells; and black dots, wild-type RAS and BRAF cells.

Close modal

Selective sensitivity of RAS- and BRAF-mutant tumor cells was the desired pharmacologic profile for RAF709 and we were encouraged to observe a relative lack of sensitivity of cells with wild-type genotypes. To further evaluate the pharmacologic profile, we compared the RAF709 sensitivity profile to RAF and MEK inhibitors that have been evaluated clinically (Fig. 5B; Supplementary Table S5). In this analysis, RAF709 profile appeared to match best with the highly selective MEK inhibitor trametinib, which showed a similar activity toward BRAF-, KRAS-, and NRAS-mutant cells and with dabrafenib, RAF709 shared sensitivity of BRAF-mutant cell lines. While the difference in potency between RAF709 and dabrafenib is striking, dabrafenib also appears to be more potent than RAF709 in a biochemical assay using purified recombinant BRAFV600E (IC50 = 1 nmol/L for RAF709 and 0.07 nmol/L for dabrafenib; ref. 25) and the relationship between biochemical and cellular potency can be impacted by multiple factors. Comparing RAF709 to sorafenib and RAF265, both relatively nonselective Type II RAF inhibitors is quite revealing. Sorafenib has no activity against RAS- or BRAF-mutant cells and RAF265 has a mixed profile, with several wild-type cells showing a similar degree of sensitivity as RAS and BRAF mutants. These data are consistent with the clinical activity observed for each compound: sorafenib's pharmacology is driven mainly by activity against angiogenic receptor tyrosine kinases (e.g., VEGFR), and while RAF265 demonstrated some activity in patients with mutant tumors, its activity is limited by what are likely off-target toxicities (15, 27).

RAF709 demonstrates antitumor activity in tumors harboring RAS mutations in vivo

Following the evaluation of RAF709 activity in human cancer cell lines, we investigated both signaling inhibition and antitumor efficacy of RAF709 in the KRASmut Calu-6 model in vivo (Fig. 6). Nude mice bearing Calu-6 xenograft tumors were treated with a single dose of RAF709 across a wide dose range (from 10 to 200 mg/kg). Tumor tissues were then collected at multiple time points postdose to determine pMEK levels. As shown in Fig. 6A, RAF709 treatment led to inhibition of MEK phosphorylation in a dose-dependent manner both in degree and in duration. RAF709 at 100 mg/kg and 200 mg/kg was able to suppress pMEK to greater than 50% for more than 16 hours. We subsequently evaluated the antitumor efficacy of RAF709 in the same tumor xenograft model. Tumor-bearing animals were dosed with vehicle, RAF709 at 10, 30, 100, or 200 mg/kg, administered orally every day (qd) for 19 days. Antitumor activity was determined by assessing percentage of tumor volume in the treatment groups versus that in vehicle-treated (% T/C) or percentage of tumor regression compared with the starting volume (% regression). In line with pMEK inhibition, RAF709 treatment resulted in dose-dependent antitumor activity starting from 30 mg/kg (Fig. 6B). Treatment with 30 mg/kg of RAF709 led to a 52% T/C, while treatment at 100 and 200 mg/kg resulted in tumor regressions of 47% and 88%, respectively, in line with the more durable pathway inhibition at the two higher dose levels. All treatment groups were well tolerated with no body weight loss at the end of the efficacy study, no signs of toxicity or mortality were observed (data not shown). Doses up to 300 mg/kg daily were also tolerated in additional studies (data not shown); however, higher doses resulted in significant body weight loss.

Figure 6.

RAF709 demonstrates antitumor activity in tumors harboring RAS mutations in vivo. A, Tumor samples were collected at the indicated time points following a single dose of vehicle or increasing doses of RAF709 in the Calu-6 tumor-bearing animals to determine pMEK levels. pMEK levels are represented as the ratio of pMEK/total MEK in the treatment group compared with vehicle control at each time point. B, Calu-6 tumor xenograft growth inhibition was measured following treatment with vehicle or RAF709 across four dose levels. Tumor volume is represented as the mean tumor volume from 6 animals per treatment group ± SEM. C,In vivo antitumor activity of RAF709 was assessed in a panel of 23 NSCLC PDX models, presented as the percentage of change in tumor volume at the time of measurement compared with initial tumor volume. Positive values indicate tumor growth and negative values indicate tumor regression. RAF709 was dosed orally once daily at 60 mg/kg or 200 mg/kg (). Each tumor is annotated for BRAF or RAS mutation status. Tumor response to paclitaxel in the same panel of PDX models was included for comparison.

Figure 6.

RAF709 demonstrates antitumor activity in tumors harboring RAS mutations in vivo. A, Tumor samples were collected at the indicated time points following a single dose of vehicle or increasing doses of RAF709 in the Calu-6 tumor-bearing animals to determine pMEK levels. pMEK levels are represented as the ratio of pMEK/total MEK in the treatment group compared with vehicle control at each time point. B, Calu-6 tumor xenograft growth inhibition was measured following treatment with vehicle or RAF709 across four dose levels. Tumor volume is represented as the mean tumor volume from 6 animals per treatment group ± SEM. C,In vivo antitumor activity of RAF709 was assessed in a panel of 23 NSCLC PDX models, presented as the percentage of change in tumor volume at the time of measurement compared with initial tumor volume. Positive values indicate tumor growth and negative values indicate tumor regression. RAF709 was dosed orally once daily at 60 mg/kg or 200 mg/kg (). Each tumor is annotated for BRAF or RAS mutation status. Tumor response to paclitaxel in the same panel of PDX models was included for comparison.

Close modal

To further assess the antitumor activity of RAF709, we performed a large-scale in vivo screen of RAF709 efficacy in 23 PDX models derived from patient non–small cell lung cancer (NSCLC) tumors. Comprehensive molecular profiling of NSCLC showed that the RAS/RAF/MEK pathway is frequently altered in this tumor type, including mutations in KRAS, NRAS, and BRAF (The Cancer Genome Atlas, 2014). We have previously reported that the NSCLC PDX models we established showed good concordance with clinical samples with respect to the genomic landscape (20). We therefore used these models to assess clinical potential of RAF709 in the treatment of NSCLC. Our screen adopted a “one animal per model per treatment” approach as described by Gao and colleagues to determine RAF709 response at the population level. Tumor response is presented as a waterfall plot of best average percentage change in tumor volume with RAF709 treatment, and tumors were annotated for their mutation status of RAS or BRAF (Fig. 6C). RAF709 dosed at 60 mg/kg or 200 mg/kg () daily led to tumor growth inhibition in a subset of NSCLC PDX tumors, in which tumors that harbor mutation of BRAF, NRAS, or KRAS were enriched among the better responders. One of the BRAFmut tumors, HLUX1323, harbors a D595N mutation that has been shown to activate signaling mediated through RAF dimerization (8). For comparison, the chemotherapeutic agent paclitaxel dosed at a clinically relevant dose (1.5 mg/kg weekly, i.v.) showed less antitumor activity across the PDX models, and its activity was not associated with BRAF/RAS mutation. These data further support that RAF709 demonstrates significant antitumor activity in tumors driven by either oncogenic BRAF or RAS.

RAF709 in combination with MEK inhibitor led to enhanced antitumor activity

RAF709 demonstrated significant activity in a subset of RAS-mutant tumors both in vitro and in vivo. To better understand the underlying biology of response, we examined the activity of RAF709 in inhibiting the pathway signaling in Calu-6 versus an insensitive model HPAF-II (IC50 = 14 μmol/L, Supplementary Fig. S4), a pancreatic adenocarcinoma cell line harboring a G12D KRAS mutation. We compared pMEK and pERK levels following treatment with RAF709 in both cell lines at 2 and 24 hours posttreatment. In Calu-6, RAF709 showed strong inhibition of MEK/ERK phosphorylation at 0.5 μmol/L at both 2 and 24 hours (Fig. 7A). In HPAF-II, RAF709 also showed inhibition of ERK phosphorylation at 0.5 μmol/L at 2 hours; however, it was unable to sustain the level of inhibition up to 24 hours at 0.5 and 5 μmol/L. Similar pathway rebound in these cells was observed with the MEK inhibitor, trametinib, where its ability to inhibit pERK was reduced at 24 hours compared with 2 hours. Pathway rebound observed with MEK inhibitors in KRAS-mutant cells has been attributed to reactivation of CRAF (10, 13). We therefore hypothesized that combining RAF709 with trametinib could lead to more sustained pathway inhibition in HPAF-II. Supporting this hypothesis, the combination of 0.5 μmol/L RAF709 and 1 nmol/L trametinib led to a stronger and more durable inhibition of pERK compared with either agent alone at the same concentration (Fig. 7A). This combination also led to induction of apoptosis as judged by the appearance of cleaved PARP (cPARP) at one tenth of the concentration required for either single agent. Following this observation, we next determined whether combination treatment also led to enhanced antiproliferative activity. Growth inhibition of HPAF-II cells was measured following treatment with RAF709 or trametinib as single agent, or with the two inhibitors in combination, across a wide dose range. Isobolograms and synergy scores were generated to assess the combination activity as described previously (19). As shown in Fig. 7B, RAF709 in combination with trametinib had a synergistic effect on inhibiting HPAF-II cell growth, with a synergy score of 11.1. These data suggest that RAF709 antitumor activity in KRAS-mutant tumors could be further enhanced by combining with a MEK inhibitor.

Figure 7.

RAF709 in combination with a MEK inhibitor provides enhanced antitumor activity in KRASmut tumors. A, The ability of RAF709 single agent to inhibit MEK/ERK phosphorylation in Calu-6 and HPAF-II, and in combination with trametinib in HPAF-II, was assessed by Western blot analysis following 2- and 24-hour treatment. Cleaved PARP was measured as a marker for cell death. B, Antiproliferative combination activity of RAF709 and trametinib in HPAF-II cells was assessed. Left, dose matrix representing percentages of growth inhibition relative to DMSO by RAF709, trametinib, and the combination following 5 days of treatment. Middle, excess inhibition values representing the deviation between the combination effect and the calculated additivity effect of the two single agents using the Loewe model. The calculated Loewe synergy score is indicated. Right, isobologram analysis of the dose matrix data. Blue line, data points; red line, additivity. The calculated Loewe combination index (CI) at 50% growth inhibition is indicated. C,In vivo activity of RAF709 and trametinib as single agents or in combination in the HPAF-II xenograft model. Left, signaling inhibition following a single dose of the treatment as measured by the DUSP6 mRNA levels. DUSP6 levels are represented as the percentage change in comparison with the vehicle group after normalization to a control gene RPLPO. In vivo tumor growth (center) and tolerability (right) following 10 days of treatment (n = 6 mice/group); treatment groups and activity are indicated.

Figure 7.

RAF709 in combination with a MEK inhibitor provides enhanced antitumor activity in KRASmut tumors. A, The ability of RAF709 single agent to inhibit MEK/ERK phosphorylation in Calu-6 and HPAF-II, and in combination with trametinib in HPAF-II, was assessed by Western blot analysis following 2- and 24-hour treatment. Cleaved PARP was measured as a marker for cell death. B, Antiproliferative combination activity of RAF709 and trametinib in HPAF-II cells was assessed. Left, dose matrix representing percentages of growth inhibition relative to DMSO by RAF709, trametinib, and the combination following 5 days of treatment. Middle, excess inhibition values representing the deviation between the combination effect and the calculated additivity effect of the two single agents using the Loewe model. The calculated Loewe synergy score is indicated. Right, isobologram analysis of the dose matrix data. Blue line, data points; red line, additivity. The calculated Loewe combination index (CI) at 50% growth inhibition is indicated. C,In vivo activity of RAF709 and trametinib as single agents or in combination in the HPAF-II xenograft model. Left, signaling inhibition following a single dose of the treatment as measured by the DUSP6 mRNA levels. DUSP6 levels are represented as the percentage change in comparison with the vehicle group after normalization to a control gene RPLPO. In vivo tumor growth (center) and tolerability (right) following 10 days of treatment (n = 6 mice/group); treatment groups and activity are indicated.

Close modal

We next assessed whether the combination activity of RAF709 and trametinib could be extended in vivo. To evaluate the effect of the combination treatment on signaling inhibition versus either agent alone, nude mice bearing HPAF-II xenograft tumors were treated with a single dose of RAF709 at 100 mg/kg, trametinib at 0.3 mg/kg or two inhibitors combined. DUSP6 mRNA levels, as a measurement of pathway activity, was determined in tumor samples collected at multiple time points postdose. As shown in Fig. 7C (left), RAF709 treatment led to 83% inhibition of DUSP6 at 4 hours postdose compared with vehicle control; however, this inhibition was not durable as demonstrated by the increased levels of DUSP6 at 16 and 24 hours postdose. Similarly, trametinib treatment led to a partial and transient inhibition of DUSP6. In contrast, the combination of RAF709 and trametinib led to a more sustained DUSP6 inhibition, showing greater than 80% of inhibition even at 16 hours postdose. We next evaluated antitumor efficacy of the different treatments in the same tumor xenograft model. Tumor-bearing animals were dosed with vehicle, RAF709 at 100 mg/kg once daily, trametinib at 0.3 mg/kg once daily, or a combination of both for 10 days. In line with DUSP6 inhibition, the combination of RAF709 and trametinib treatment resulted in greater antitumor activity than either of the single agents alone, resulting in 33% regression as compared with 40% T/C or 54% T/C by RAF709 or trametinib, respectively (Fig. 7C, center). It should be noted, however, that this combination dose and regimen may not be tolerated long-term since one mouse had to be euthanized due to dose-limiting body weight loss and the average body weight loss in the group was 10% (right). These data support that combination of RAF709 with MEK inhibitor trametinib could lead to more durable pathway inhibition and higher antitumor activity; however, alternative dose schedules will need to be adopted for longer term treatment.

Despite the clinical activity demonstrated by RAF and MEK inhibitors in BRAFV600-mutant melanoma, these inhibitors are relatively ineffective in mutant RAS-driven cancers, tumors in which there remains a high unmet clinical need. Recent advances have been made in developing drugs against the KRAS G12C–mutant protein directly through covalent inhibition (3, 4). However, development of this class of molecules into potential therapeutics would require further improvement in cellular potency and exploration of combinations that could alleviate receptor tyrosine kinase (RTK) activation and/or downstream pathway reactivation. Inhibition of RAF downstream of mutant RAS has been a long-standing goal; however, RAF inhibitors such as dabrafenib and vemurafenib, induce paradoxical activation in RASmut tumors and this has been attributed to their reduced affinity for the second site of BRAF-CRAF dimer after occupancy of the first protomer site, leading to increased RAF dimerization and transactivation (8). The structural basis for the differential effects of RAF inhibitors has recently been described by Karoulia and colleagues (28). The model they developed based partially on X-ray cocrystal structures with BRAF and a set of ATP-competitive RAF inhibitors, indicates that the pharmacology is driven by a combination of drug-induced dimerization, which is dependent on the position of the alpha-C helix, and the potentiation of RAF/RAS-GTP and RAF/MEK interactions by the inhibitor.

The development of so-called pan-RAF inhibitors with the goal of treating RAS-mutant tumors has been described in two recent reports; however, in addition to inhibiting RAF isoforms, these compounds also exhibit activity against many other kinases. For example, the ability to target SRC signaling was attributed to the antitumor efficacy of CCT196969 and CCT241161 in drug-resistant BRAFmut melanomas with upregulated RTK signaling (29). However, as with LY3009120, RAF265, BGB-283, and sorafenib, the spectrum of off-target activities of these inhibitors may impact their ability to potently inhibit RAF at tolerated doses. To address the need for a highly selective RAF inhibitor with the ability to inhibit pathway signaling in RASmut tumors, we have developed RAF709, an inhibitor with low to subnanomolar potency targeting RAF isoforms and a high level of selectivity against other kinases. Similar to the pan-RAF inhibitor LY300912, RAF709 exhibits a type II binding mode in the RAF ATP pocket with the DFG loop out and αC helix-in conformation. Consistent with other ATP-competitive inhibitors, RAF709 stabilizes B/CRAF dimerization; however, unlike other RAF inhibitors, it demonstrates equipotency at inhibiting both RAF monomer- and dimer-driven signaling in cellular assays. The ability of RAF709 to effectively inhibit RAF dimer activity leads to its minimal paradoxical activation and antitumor activity in not only BRAFV600 but also RASmut tumors. RAF709 demonstrates enriched activity in BRAF- and RAS-mutant tumors in a large in vitro cancer cell line panel and in in vivo PDX screens.

In addition to the most prevalent BRAF V600 mutations, other BRAF mutations have been identified across different tumor types. Studies of NSCLC clinical samples revealed that non-V600 BRAF mutations represented 50%–75% of all BRAF mutations (30, 31). These non-V600 mutations have been experimentally characterized as activating or kinase-impaired, and all could activate pathway signaling in a CRAF dimerization–dependent manner (8, 32, 33). The ability of RAF709 to inhibit RAF dimers suggests that it would be active in tumors that harbor these BRAF mutations. Indeed RAF709 has shown antiproliferative activity in cancer cell lines that harbor non-V600 mutation with or without a cooccurring KRAS mutation (Supplementary Table S5), including Hey-A8 (KRASG12D/BRAFG464E, IC50 = 0.63 μmol/L), MDA-MB-231 (KRASG13D/BRAFG464V, IC50 = 3.4 μmol/L), and NCI-H1666 (BRAFG466V, IC50 = 3.97 μmol/L). In comparison, the BRAF monomer inhibitor dabrafenib had IC50>30 μmol/L in all three cell line models. In addition, RAF709 showed in vivo antitumor activity in HLUX1323, a NSCLC PDX model that harbors BRAFD594N mutation, leading to 26% of tumor shrinkage (Fig. 6). Our data suggest that RAF709 could be effective for the treatment of tumors driven by atypical BRAF mutations that do not respond to the current BRAFV600 inhibitors.

Data from our cell line and PDX panels suggest that tumors with BRAF or RAS mutations are more sensitive to RAF709 treatment. Within these molecular subtypes, however, sustained pathway inhibition is required for optimal antitumor activity. This hypothesis is supported by our findings that combination of RAF709 and trametinib led to more durable pathway inhibition and anticancer activity compared with either agent alone. These data are consistent with reports from the functional genomic screens that demonstrated that targeting CRAF could sensitize response to MEK inhibitors in KRAS-mutant tumors (10, 13). In addition to targeting multiple signaling nodes within the MAPK pathway to maximize both level and duration of signaling inhibition, combining RAF709 with inhibitors targeting the upstream RTK activation such as EGFR (34, 35), bypass mechanisms such as YAP1 (36) or targeting prosurvival BCL2 family members (37), may further enhance efficacy and reduce the development of drug resistance.

Thus, we have developed a highly selective pan-RAF inhibitor that is equipotent at inhibiting RAF monomers and dimers. As a result, RAF709 demonstrates antitumor activity in both BRAF- and RAS-mutant tumors. RAF709 is orally bioavailable and exhibits a pharmacodynamic/efficacy relationship, which indicates that sustained target inhibition leads to tumor regression in vivo. The activity of RAF709 in KRASmut tumors could be further enhanced by combination with a MEK inhibitor. The mouse xenograft studies described here provide only a rough estimate of the therapeutic index of RAF709 and detailed histopathologic studies have not been conducted. However, highly selective RAF dimer inhibitors like RAF709 provide the opportunity to effectively test the therapeutic potential of RAF inhibition in atypical BRAF-mutant and RAS-mutant tumors, without the risk of inhibiting off-target kinases as observed with previous inhibitors in this class. Ongoing clinical trials with LXH254, which is structurally related to RAF709, will answer this critical question.

N. Keen is a chief scientific officer and has a commercial research grant from Bicycle Therapeutics Ltd. M.P. Dillon has ownership interest (including patents) in Novartis. W.R. Sellers has ownership interest (including patents) in Novartis Pharmaceuticals and Peloton Pharmaceuticals, is a consultant/advisory board member for Sanofi Pharmaceuticals, Servier Pharmaceuticals, Astex Pharmaceuticals, Merck-Serono Pharmaceuticals, and Ideya Pharmaceuticals. D.D. Stuart is a director at Novartis and has ownership interest (including patents) in Novartis. No potential conflicts of interest were disclosed by the other authors.

Conception and design: W. Shao, Y. Feng, G. Nishiguchi, J. Tellew, J. Haling, V.R. Polyakov, R. Zang, P. Amiri, M. Singh, N. Keen, M.P. Dillon, E. Lees, S. Ramurthy, D.D. Stuart

Development of methodology: Y.M. Mishina, Y. Feng, J. Haling, R. Aversa, R. Zang, P. Amiri, D.D. Stuart

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): Y. Feng, V.G. Cooke, S. Rivera, Y. Wang, F. Shen, L.A. Mathews Griner, R. Zang, M. Singh, S. Ramurthy, D.D. Stuart

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): W. Shao, Y.M. Mishina, Y. Feng, G. Caponigro, Y. Wang, F. Shen, J.M. Korn, L.A. Mathews Griner, J. Haling, V.R. Polyakov, R. Zang, M. Hekmat-Nejad, P. Amiri, M. Singh, N. Keen, M.P. Dillon, E. Lees, S. Ramurthy, D.D. Stuart

Writing, review, and/or revision of the manuscript: W. Shao, Y.M. Mishina, Y. Feng, G. Caponigro, V.G. Cooke, A. Rico, J. Haling, V.R. Polyakov, M. Singh, E. Lees, W.R. Sellers, D.D. Stuart

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): Y.M. Mishina, Y. Feng, G. Caponigro, S. Rivera, J. Tellew, R. Zang, M. Singh, S. Ramurthy, D.D. Stuart

Study supervision:W. Shao, V.G. Cooke, R. Zang, M. Singh, E. Lees, S. Ramurthy, W.R. Sellers, D.D. Stuart

The authors thank Brent Appleton, Ann Van Abbema, Laura Tandeske, Hanne Merritt, Sylvia Ma, and Hanneke Jansen for scientific and technical support during the early discovery phase of this project. We would also like to acknowledge Novartis Institutes for Biomedical Research for funding.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Stephen
AG
,
Esposito
D
,
Bagni
RK
,
Mccormick
F
. 
Dragging ras back in the ring
.
Cancer Cell
2014
;
25
:
272
81
.
2.
Ostrem
JM
,
Peters
U
,
Sos
ML
,
Wells
JA
,
Shokat
KM
. 
K-Ras(G12C) inhibitors allosterically control GTP affinity and effector interactions
.
Nature
2013
;
503
:
548
51
.
3.
Lito
P
,
Solomon
M
,
Li
LS
,
Hansen
R
,
Rosen
N
. 
Allele-specific inhibitors inactivate mutant KRAS G12C by a trapping mechanism
.
Science
2016
;
351
:
604
8
.
4.
Patricelli
MP
,
Janes
MR
,
Li
LS
,
Hansen
R
,
Peters
U
,
Kessler
LV
, et al
Selective inhibition of oncogenic KRAS output with small molecules targeting the inactive state
.
Cancer Discov
2016
;
6
:
316
29
.
5.
Hatzivassiliou
G
,
Song
K
,
Yen
I
,
Brandhuber
BJ
,
Anderson
DJ
,
Alvarado
R
, et al
RAF inhibitors prime wild-type RAF to activate the MAPK pathway and enhance growth
.
Nature
2010
;
464
:
431
5
.
6.
Heidorn
SJ
,
Milagre
C
,
Whittaker
S
,
Nourry
A
,
Niculescu-Duvas
I
,
Dhomen
N
, et al
Kinase-dead BRAF and oncogenic RAS cooperate to drive tumor progression through CRAF
.
Cell
2010
;
140
:
209
21
.
7.
Poulikakos
P
,
Zhang
C
,
Bollag
G
,
Shokat
K
,
Rosen
N
. 
RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF
.
Nature
2010
;
464
:
427
30
.
8.
Yao
Z
,
Torres
NM
,
Tao
A
,
Gao
Y
,
Luo
L
,
Li
Q
, et al
BRAF mutants evade ERK-dependent feedback by different mechanisms that determine their sensitivity to pharmacologic inhibition
.
Cancer Cell
2015
;
28
:
370
83
.
9.
Caunt
CJ
,
Sale
MJ
,
Smith
PD
,
Cook
SJ
. 
MEK1 and MEK2 inhibitors and cancer therapy: the long and winding road
.
Nat Rev Cancer
2015
;
15
:
577
92
.
10.
Lito
P
,
Saborowski
A
,
Yue
J
,
Solomon
M
,
Joseph
E
,
Gadal
S
, et al
Disruption of CRAF-mediated MEK activation is required for effective MEK inhibition in KRAS mutant tumors
.
Cancer Cell
2014
;
25
:
697
710
.
11.
Blasco
RB
,
Francoz
S
,
Santamaria
D
,
Canamero
M
,
Dubus
P
,
Charron
J
, et al
c-Raf, but not B-Raf, is essential for development of K-Ras oncogene-driven non-small cell lung carcinoma
.
Cancer Cell
2011
;
19
:
652
63
.
12.
Karreth
FA
,
Frese
KK
,
Denicola
GM
,
Baccarini
M
,
Tuveson
DA
. 
C-Raf is required for the initiation of lung cancer by K-Ras(G12D)
.
Cancer Discov
2011
;
1
:
128
36
.
13.
Lamba
S
,
Russo
M
,
Sun
C
,
Lazzari
L
,
Cancelliere
C
,
Grernrum
W
, et al
RAF suppression synergizes with MEK inhibition in KRAS mutant cancer cells
.
Cell Rep
2014
;
8
:
1475
83
.
14.
Peng
SB
,
Henry
JR
,
Kaufman
MD
,
Lu
WP
,
Smith
BD
,
Vogeti
S
, et al
Inhibition of RAF isoforms and active dimers by LY3009120 leads to anti-tumor activities in RAS or BRAF mutant cancers
.
Cancer Cell
2015
;
28
:
384
98
.
15.
Izar
B
,
Sharfman
W
,
Hodi
FS
,
Lawrence
D
,
Flaherty
KT
,
Amaravadi
R
, et al
A first-in-human phase I, multicenter, open-label, dose-escalation study of the oral RAF/VEGFR-2 inhibitor (RAF265) in locally advanced or metastatic melanoma independent from BRAF mutation status
.
Cancer Med
2017
;
6
:
1904
14
.
16.
Desai
J
,
Gan
H
,
Barrow
C
,
Jameson
MB
,
Mcarthur
G
,
Tran
B
, et al
Abstract CT005: phase I study of RAF dimer inhibitor BGB-283 in patients with B-RAF or K-RAS/N-RAS mutated solid tumors
.
Cancer Res
2016
;
76
:
CT005
.
17.
Tang
Z
,
Yuan
X
,
Du
R
,
Cheung
SH
,
Zhang
G
,
Wei
J
, et al
BGB-283, a novel RAF kinase and EGFR inhibitor, displays potent antitumor activity in BRAF-mutated colorectal cancers
.
Mol Cancer Ther
2015
;
14
:
2187
97
.
18.
Barretina
J
,
Caponigro
G
,
Stransky
N
,
Venkatesan
K
,
Margolin
AA
,
Kim
S
, et al
The cancer cell line encyclopedia enables predictive modelling of anticancer drug sensitivity
.
Nature
2012
;
483
:
603
307
.
19.
Lehar
J
,
Krueger
AS
,
Avery
W
,
Heilbut
AM
,
Johansen
LM
,
Price
ER
, et al
Synergistic drug combinations tend to improve therapeutically relevant selectivity
.
Nat Biotechnol
2009
;
27
:
659
66
.
20.
Gao
H
,
Korn
JM
,
Ferretti
S
,
Monahan
JE
,
Wang
Y
,
Singh
M
, et al
High-throughput screening using patient-derived tumor xenografts to predict clinical trial drug response
.
Nat Med
2015
;
21
:
1318
25
.
21.
Nishiguchi
G
,
Rico
A
,
Tanner
H
,
Aversa
R
,
Taft
B
,
S
S
, et al
Design and Discovery of N-(2-Methyl-5′-morpholino-6′-((tetrahydro-2H-pyran-4-yl)oxy)-[3,3′-bipyridin]-5-yl)-3-(trifluoromethyl)benzamide (RAF709): a potent, selective, and efficacious RAF inhibitor targeting RAS mutant cancers
.
J Med Chem
2017
;
60
:
4869
81
.
22.
Fabian
MA
,
Biggs
WH
 III
,
Treiber
DK
,
Atteridge
CE
,
Azimioara
MD
,
Benedetti
MG
, et al
A small molecule-kinase interaction map for clinical kinase inhibitors
.
Nat Biotechnol
2005
;
23
:
329
36
.
23.
Liu
Q
,
Kirubakaran
S
,
Hur
W
,
Niepel
M
,
Westover
K
,
Thoreen
CC
, et al
Kinome-wide selectivity profiling of ATP-competitive mammalian target of rapamycin (mTOR) inhibitors and characterization of their binding kinetics
.
J Biol Chem
2012
;
287
:
9742
52
.
24.
Warmuth
M
,
Kim
S
,
Gu
XJ
,
Xia
G
,
Adrian
F
. 
Ba/F3 cells and their use in kinase drug discovery
.
Curr Opin Oncol
2007
;
19
:
55
60
.
25.
Delord
JP
,
Robert
C
,
Nyakas
M
,
Mcarthur
GA
,
Kudchakar
R
,
Mahipal
A
, et al
Phase I dose-escalation and -expansion study of the BRAF inhibitor encorafenib (LGX818) in metastatic BRAF-mutant melanoma
.
Clin Cancer Res
2017
;
23
:
5339
48
.
26.
Poulikakos
PI
,
Persaud
Y
,
Janakiraman
M
,
Kong
X
,
Ng
C
,
Moriceau
G
, et al
RAF inhibitor resistance is mediated by dimerization of aberrantly spliced BRAF(V600E)
.
Nature
2011
;
480
:
387
390
.
27.
Strumberg
D
,
Richly
H
,
Hilger
RA
,
Schleucher
N
,
Korfee
S
,
Tewes
M
, et al
Phase I clinical and pharmacokinetic study of the novel raf kinase and vascular endothelial growth factor receptor inhibitor BAY 43–9006 in patients with advanced refractory solid tumors
.
J Clin Oncol
2005
;
23
:
965
72
.
28.
Karoulia
Z
,
Wu
Y
,
Ahmed
TA
,
Xin
Q
,
Bollard
J
,
Krepler
C
, et al
An integrated model of RAF inhibitor action predicts inhibitor activity against oncogenic BRAF signaling
.
Cancer Cell
2016
;
30
:
485
98
.
29.
Girotti
MR
,
Lopes
F
,
Preece
N
,
Niculescu-Duvaz
D
,
Zambon
A
,
Davies
L
, et al
Paradox-breaking RAF inhibitors that also target SRC are effective in drug-resistant BRAF mutant melanoma
.
Cancer Cell
2015
;
27
:
85
96
.
30.
Paik
PK
,
Arcila
ME
,
Fara
M
,
Sima
CS
,
Miller
VA
,
Kris
MG
, et al
Clinical characteristics of patients with lung adenocarcinomas harboring BRAF mutations
.
J Clin Oncol
2011
;
29
:
2046
51
.
31.
Noeparast
A
,
Teugels
E
,
Giron
P
,
Verschelden
G
,
De Brakeleer
S
,
Decoster
L
, et al
Non-V600 BRAF mutations recurrently found in lung cancer predict sensitivity to the combination of Trametinib and Dabrafenib
.
Oncotarget
2016
;
8
:
60094
108
.
32.
Freeman
AK
,
Ritt
DA
,
Morrison
DK
. 
Effects of Raf dimerization and its inhibition on normal and disease-associated Raf signaling
.
Mol Cell
2013
;
49
:
751
8
.
33.
Wan
PT
,
Garnett
MJ
,
Roe
SM
,
Lee
S
,
Niculescu-Duvaz
D
,
Good
VM
, et al
Mechanism of activation of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF
.
Cell
2004
;
116
:
855
67
.
34.
Corcoran
RB
,
Ebi
H
,
Turke
AB
,
Coffee
EM
,
Nishino
M
,
Cogdill
AP
, et al
EGFR-mediated reactivation of MAPK signaling contributes to insensitivity of BRAF-mutant colorectal cancers to RAF INHIBITION with Vemurafenib
.
Cancer Discov
2012
;
2
:
227
35
.
35.
Prahallad
A
,
Sun
C
,
Huang
S
,
Di Nicolantonio
F
,
Salazar
R
,
Zecchin
D
, et al
Unresponsiveness of colon cancer to BRAF(V600E) inhibition through feedback activation of EGFR
.
Nature
2012
;
483
:
100
13
.
36.
Lin
L
,
Bivona
TG
. 
The Hippo effector YAP regulates the response of cancer cells to MAPK pathway inhibitors
.
Mol Cell Oncol
2016
;
3
:
e1021441
.
37.
Okumura
S
,
Janne
PA
. 
Molecular pathways: the basis for rational combination using MEK inhibitors in KRAS-mutant cancers
.
Clin Cancer Res
2014
;
20
:
4193
9
.