LXH254, a Potent and Selective ARAF-Sparing Inhibitor of BRAF and CRAF for the Treatment of MAPK-Driven Tumors

Author Manuscript Published OnlineFirst on December 22, 2020; DOI: 10.1158/1078-0432.CCR-20-2563 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

LXH254, a potent and selective ARAF-sparing inhibitor of BRAF and CRAF for the treatment of MAPK-driven tumors

Kelli-Ann Monaco2, Scott Delach1, Jing Yuan1, Yuji Mishina1, Paul Fordjour1, Emma Labrot1, Daniel McKay5, Ribo Guo1 , Stacy Rivera1 , Hui Qin Wang1, Jinsheng Liang1, Karen Bui1, John Green1, Peter Aspesi1, Jessi Ambrose1, Penny Mapa1, Lesley Griner1, Mariela Jaskelioff3, John Fuller4, Kenneth Crawford6, Gwynn Pardee4, Stephania Widger4, Peter S. Hammerman1, Jeffrey A. Engelman1, Darrin D. Stuart7, Vesselina Cooke1, Giordano Caponigro1

1. Novartis Institutes for Biomedical Research, 250 Massachusetts Ave. Cambridge, MA, 02139 United States 2. Bristol Myers-Squibb, 100 Binney St, Cambridge, MA 02142, United States 3. Skyhawk Therapeutics, 35 Gatehouse Dr. Waltham, MA, 02451 United States 4. Novartis Institutes for Biomedical Research, 5300 Chiron Way, Emeryville, CA, 94608 United States 5. Ventus Therapeutics, 7150 Frederick-Banting, Montreal, Quebec H9S 2A1 Canada 6. AbSci 101 East 6th St. #350 Vancouver WA, 98660, United States 7. Scorpion Therapeutics 888 Boylston St. Suite 1111 Boston MA, 02199, United States

Running Title: LXH254 a potent and selective BRAF and CRAF inhibitor

Corresponding author: Giordano Caponigro, Novartis Institutes for Biomedical Research, 250 Massachusetts Avenue, Cambridge MA 02139. Phone 617-871-7395, Fax: 671-871-4213; Email: [email protected]

Downloaded from clincancerres.aacrjournals.org on October 6, 2021. © 2020 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 22, 2020; DOI: 10.1158/1078-0432.CCR-20-2563 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Translational Relevance

The RAS/RAF/MEK/ERK pathway is activated in greater than 30% of human cancers primarily via mutations in KRAS, NRAS and BRAF. The development of highly selective and genetic context-specific RAF inhibitors has led to spectacular improvements in patients with BRAFV600-mutant disease. Unfortunately, the BRAF V600E selectivity of these inhibitors prevents their utility in RAS-mutant tumors. Here, we describe an unexpected paralog-selectivity for the type II RAF inhibitor LXH254 that is currently undergoing clinical trials in RAS-driven tumors in combination with MEK and ERK inhibitors. LXH254 is a potent inhibitor of both monomeric and dimeric BRAF and

CRAF, but a comparably poor inhibitor of ARAF. This paralog selectivity permits robust inhibition of dimeric BRAF and CRAF simultaneous with potentially improved TI.

This tolerability profile has enabled vertical combinations with MEK and ERK inhibitors for the treatment of RAS–driven tumors in clinic.

Abstract

Purpose: Targeting RAF for anti-tumor therapy in RAS-mutant tumors holds promise.

Herein we describe in detail novel properties of the type II RAF inhibitor LXH254.

Experimental Design: LXH254 was profiled in biochemical, in vitro, and in vivo assays including examining the activities of the drug in a large panel of cancer-derived cell lines, and a comprehensive set of in vivo models. In addition, activity of LXH254 was assessed in cells where different sets of RAF paralogs were ablated, or that expressed kinase-impaired and dimer-deficient variants of ARAF.

Results: We describe an unexpected paralog selectivity of LXH254, which is able to potently inhibit BRAF and CRAF, but has less activity against ARAF. LXH254 was

active in models harboring BRAF alterations, including atypical BRAF alterations co-

expressed with mutant K/NRAS, and NRAS mutants, but had only modest activity in

KRAS mutants. In RAS mutant lines loss of ARAF, but not BRAF or CRAF, sensitized

cells to LXH254. ARAF-mediated resistance to LXH254 required both kinase function

and dimerization. Higher concentrations of LXH254 were required to inhibit signaling in

RAS-mutant cells expressing only ARAF relative to BRAF or CRAF. Moreover,

specifically in cells expressing only ARAF, LXH254 caused paradoxical activation of

MAPK signaling in a manner similar to dabrafenib. Lastly, in vivo, LXH254 drove complete regressions of isogenic variants of RAS mutant cells lacking ARAF expression, while parental lines were only modestly sensitive.

Conclusions: LXH254 is a novel RAF-inhibitor able to inhibit dimerized BRAF and

CRAF as well as monomeric BRAF while largely sparing ARAF.

Introduction

Activation of the RAS/RAF pathway due to oncogenic lesions occurs in more than a

third of human cancers, most commonly via mutations in KRAS, NRAS and BRAF (1,2).

Not surprisingly, significant effort has gone into developing inhibitors targeting different

nodes in this pathway most notably the RAF, MEK and ERK kinases and more recently

the KRAS G12C variant (3-7).

The requirement for RAS/RAF signaling in normal tissues limits the potential therapeutic

index (TI) of RAF/MEK/ERK inhibitors. However, studies in mice and humans indicate

that selective inhibition of RAF paralogs and/or oncogenic variants may enable the

development of inhibitors with favorable TIs. First, in adult mice, combined deletion of

BRAF and CRAF is tolerated, whereas combined removal of either MEK1/MEK2, or

ERK1/ERK2 is lethal (8). Second, RAF inhibitors selective for V600-genetic variants

have been developed, and tellingly these inhibitors have proven far more effective

clinically than have MEK inhibitors which inhibit equally both MEK1 and MEK2 (9-19).

The simplest interpretation of differing clinical results for these two inhibitor classes is that mutant selectivity of the BRAF inhibitors confers a favorable TI enabling relatively

greater systemic drug exposure and correspondingly improved pathway suppression in

tumors. Thus, agents that selectively target specific RAF isoforms or genetic variants

should have improved TIs relative to agents that inhibit equally all family members of

any given node in the pathway.

The mutant-selectivity of BRAFV600E/D/K inhibitors precludes their use in KRAS, NRAS

and atypical BRAF mutant tumors. Unfortunately, MEK inhibitors, which effectively inhibit signaling by mutant RAS in pre-clinical models, have proven largely ineffective in

RAS-mutant tumors clinically (20). This failure likely results from two non-mutually exclusive sources. First, RAS activates multiple downstream pathways. Thus, although

RAF/MEK/ERK is a major RAS effector pathway, in many tumors additional pathways

may be the major oncogenic driver, or provide redundancy should RAF signaling be

impaired (21). Second, the intricate feedback mechanisms that serve to control RAF

pathway output in normal tissues, while disabled in BRAFV600 mutant disease (22), are

largely intact in RAS-mutant tumors. Thus, RAS-mutant tumors contain evolutionarily

honed mechanisms that can buffer tumor cells against the effects of RAF-pathway

inhibitors (21,23). This feedback effect, coupled with the relatively low systemic

exposures afforded by the narrow TI, likely places severe constraints on potential single

agent activity of such inhibitors. Inhibitors that impair RAF-pathway signaling with at

least some degree of bias for the RAS-mutant context relative to the wild type setting will be required to test the dependency of mutant RAS on RAF/MEK/ERK signaling in clinic.

Herein we describe LXH254, a RAF inhibitor with potent activity against both BRAF and

CRAF but 30-50 fold lower activity against ARAF in cells. LXH254 was predominantly

active in tumor models harboring activating variants of BRAF and NRAS, but less so in

KRAS mutants, with the exception of KRAS mutants harboring co-incident mutations in

BRAF. Critically, ablation of ARAF expression combined with LXH254 to elicit profound

anti-proliferative in vitro and anti-tumor activity in vivo. Collectively these data suggest

that LXH254 exemplifies a novel class of RAF inhibitors with selectivity for specific RAF

paralogs. Reduced inhibitory activity against ARAF may limit LXH254 single agent

activity to either tumors within specific genetic contexts such as BRAF mutant tumors, or

RAS-mutant tumors positive for biomarkers indicative of a lack of ARAF utilization.

However, partial pathway inhibition provided by LXH254 in many tumor cells due to potent inhibition of BRAF and CRAF, coupled with improved tolerability in normal tissues due to ARAF avoidance, may enable tolerable combinations with additional agents targeting downstream effectors such as MEK, ERK, or CDK4/6.

Methods:

Cell culture

HCT 116 RAF single and double knockout lines were generated using parental and BRAF-/- cells from Horizon Discovery (Cambridge, United Kingdom). All other cells lines were obtained from the BROAD/Novartis Cell Line Encyclopedia collection (24) and maintained in Dulbecco's Modified Eagle Medium (MIA PaCa-2, 293T, HEK-293), McCoy’s 5A (HCT 116), Eagle's minimal essential medium (Calu-6, HeyA8, SK-MEL-2) or RPMI (all other lines) supplemented with 10% fetal bovine serum (FBS, VWR Life Sciences Seradigm #1500-500). Lines were confirmed mycoplasma-free by PCR and cell line identity was confirmed by SNP genotyping. All cell lines were maintained at 37°C in a humidified atmosphere containing 5% carbon dioxide.

RAF in vitro enzyme assays Biochemical inhibition of ARAF, BRAF and CRAF shown in Fig. 1A was performed as described for RAF709 in (25). In brief, the catalytically inactive MEK1K97R variant was used as a substrate for either full-length BRAF, full-length activated CRAF

(Y340E/Y341E variant), or a full-length N-terminal GST-ARAF fusion. In all cases, LXH254 was pre-incubated with CRAF/BRAF/ARAF for 30 minutes prior to substrate addition/reaction initiation. Biochemical activity of LXH254 in the extended panel of kinases shown in supplementary table 1 was determined as described in (26).

Generation of knock-out cell lines

To generate RAF knockout lines, clustered regularly interspersed short palindromic repeat (CRISPR) genome editing (27,28) was used for all lines except HCT 116. Guide sequences for ARAF (CCTGCCCAACAAGCAACGCA), BRAF (ATATCAGTGTCCCAACCAAT) and CRAF (TGTTGCAGTAAAGATCCTAA) were synthesized using standard chemistry and cloned into in-house generated lentiviral expression vectors. . Lentiviral vectors were packaged in 293T cells with TransIT-293 Transfection Reagent (Mirus #MIR2700) using 5.4 micrograms of plasmid per 10-cm plate. Virus was harvested after 48 hours and passed through a 0.45 micron filter. Cell lines stably expressing Cas9 were incubated with lentivirus and 8 µg/ml polybrene (Millipore #TR-1003-G) for 24 hours, and subsequently cultured in media containing puromycin. Following selection in puromycin reductions in target protein levels were assessed via immunoblotting and single clones with complete loss of target protein expression isolated from pools via serial dilutions. HCT 116 knock out lines were generated in parental and BRAF-/- lines from Horizon Discovery by transfection with zinc finger mRNAs against ARAF, BRAF and CRAF purchased from Sigma.

Generation of cell lines expressing exogenous ARAF constructs

ARAFWT-FLAG, ARAFD447A-FLAG, ARAFK336M-FLAG and ARAFR362H-FLAG cDNA constructs were synthesized and cloned into in-house generated lentiviral vectors and expressed from either a constitutive (pXP1704) or doxycycline-inducible (pXP1509A) EF1-alpha promoter by GeneArt (Thermo Fisher Scientific). Lentivirus was packaged, harvested and used to infect cells as previously described. Cells were cultured in media containing G418 MIA PaCa-2 until stable pools were generated. Protein expression was confirmed by immunoblotting and mutations by MassArray (Agena Biosciences).

Cell Proliferation Assays

Cells were seeded in 96-well plates (Corning #3903) and 24 hours later compound dilution series (1:3 starting at 10M) were added to wells in duplicate or triplicate. All compounds were synthesized at Novartis Pharma AG. Plates were incubated for 72 - 120 hours and then CellTiter-GloTM (CTG) Luminescent Cell Viability Assay (Promega #G7573) was added to wells to assess ATP levels. All points were normalized to

DMSO control samples when determining IC50 values and calculations performed using GraphPad PRISM 7.0.

High-throughput profiling of the cell line panel described in supplemental figure 1A was performed as described in Barretina et al (24). pMEK1/2 and pERK1/2 MSD Assays For Meso Scale Discoveries (MSD) assays, cells were incubated with compound for 4 or 24 hrs at 37°C and subsequently lysed for 30 minutes at 4°C. Assays were performed as described in the manufacturer’s protocol for Phospho/Total ERK1/2 (MSD #K15107D-1), phospho-protein levels normalized to DMSO-treated parental samples and fold change calculated based upon percentage phosphoprotein in each sample as

described in the manufacturer’s protocol. Washout experiments described in Figure 1C were performed as described in (25).

Protein Immunoblots

Cells were plated in 6-well dishes (Corning #3506) and treated with inhibitors for 4 to 72 hours. Cells were harvested in RIPA Lysis and Extraction Buffer (Thermo Fisher Scientific # 89900) containing 1X Halt Protease Inhibitor Cocktail (Thermo Fisher Scientific # 87785) and Halt Phosphatase Inhibitor Cocktail (Thermo Fisher Scientific # 78420). Lysates were used for western blot analysis with antibodies recognizing ARAF (Cell Signaling #4432), BRAF (Sigma-Aldrich # HPA00132), CRAF (Cell Signaling #12552), CRAF (BD Biosciences # 610151), FLAG (Cell Signaling #8146), phospho- MEK1/2 (S217/S221) (Cell Signaling #9154), phospho-ERK1/2 (T202/Y204) (Cell Signaling #4370) and GAPDH (Millipore #MAB374) diluted 1:1000 or with β-actin antibody (Life Technologies #AM4302) diluted 1:5000.

Co-immunoprecipitations For Immunoprecipitations shown in Fig. 1B, Cells were seeded in 15-cm dishes and incubated with 1M of the indicated compounds for 1 hour. Cell lysates were prepared in immunoprecipitation buffer (10 mM Tris pH7.4, 50 mM NaCl, 0.5% (v/v) NP-40, 1 mM EDTA) supplemented with 1× protease and 1× phosphatase inhibitor cocktails. Cleared lysates were normalized for protein concentration and incubated with BRAF (Sigma # HPA001328), CRAF (Bethyl Labs (A301-519A) (HCT 116), Millipore # 04-739) or MEK (Millipore # 07-641) antibodies as indicated overnight at 4°C. Protein A Ultra Link Resin (Thermo Scientific) was then added to each sample, incubated for 2 hrs at 4°C, washed with immunoprecipitation lysis buffer and eluted in SDS sample buffer. In all other immunoprecipitations, cells were treated +/- compound for 2-24 hrs, cells lysed in Pierce IP Lysis Buffer (Thermo Fisher #87787) containing 1X Protease (Thermo Fisher Scientific # 87785) and Phosphatase Inhibitor (Thermo Fisher Scientific # 78420) cocktails, and lysates clarified by centrifugation. For each sample, 1-2 milligrams of lysate were incubated with antibodies to ARAF (Cell Signaling #4432), BRAF (Sigma- Aldrich # HPA00132), or FLAG (Sigma-Aldrich # F7425) and Protein G Sepharose

beads (GE Healthcare #17-0618-01) for 3 hrs at 4°C. Complexes were washed 3x with IP Lysis Buffer and analyzed by immunoblotting.

Cell-based kinase assays In-cell kinase selectivity profiling was performed by KiNativ™. Briefly, HCT 116 cells were treated with LXH254 for 2 hours at 10 µM and then lysates were processed, probe-labelled and analyzed by LC-MS/MS as previously described (29). For IP-kinase assays, cells were cultured in compound for 4-24 hrs prior to protein immunoprecipitation as described above. Immunoprecipitates were washed once with IP Lysis buffer, twice with 1X Kinase buffer (Cell Signaling #9802) and protein-bound beads incubated in 2X Kinase buffer with 250 micromolar ATP (Cell Signaling #9804) and 0.5 micrograms MEK1 (Millipore #14-420) or MEK1-K97M (Proqinase #0785-0000- 1) for 30 minutes at 30°C. Beads were then washed 3X with IP-lysis buffer and prepared for immunoblotting as described above.

In vivo studies

Mice and statement of welfare:

Outbred athymic (nu/nu) female mice (“HSD: Athymic Nude-nu”) (Charles River) and

SCID beige (Charles River) mice were allowed to acclimate in the Novartis NIBRI animal facility with access to food and water ad libitum for minimum of 3 days prior to manipulation. Animals were handled in accordance with Novartis ACUC regulations and guidelines.

Cell culture and in vivo efficacy

All cell lines were shown to be free of Mycoplasma sp. and murine viruses in the

IMPACT VIII PCR assay panel (IDEXX RADIL, IDEXX Laboratories INC, Westbrook,

ME). In all cases cells were harvested at 80-95% confluence with 0.25% trypsin-EDTA

(Gibco #25200-056), washed with PBS, detached with 0.25% trypsin-EDTA (Gibco

#25200-056) and neutralized with growth medium. Following centrifugation for 5 min at

1200 rpm, all cells except HEY-A8 cells (PBS) were re-suspended in either cold HBSS

(Gibco #14175-095) (HCT 116, A375) or cold HBSS and an equal volume of Matrigel™

Matrix (Corning #354234, all other cells). Cells were then injected in 100-200 l

volumes into either the right or left flanks of female nude mice with the exception of

MEL-JUSO cells, which were injected into SCID beige mice. Total cell numbers

injected/mouse were 2x106 (HCT 116, HEY-A8), 5x106 (MIA PaCa-2, HPAF-II, A375,

Hs944.T, SK-MEL-30), or 1x107 (MEL JUSO, Calu-6, PC-3). 5-7 mice were used for

each group.

3990HPAX, 2043HPAX, 1290HCOX, 2861HCOX, 1855HCOX, 2094HLUX, and

3486HMEX patient-derived tumor xenograft (PDX) tumors and the WM793 xenograft

were propagated by serial passage of tumor fragments in nude mice. Briefly, 3x3x3 mm

fragments of fresh tumor from a previous passage were implanted subcutaneously into

the mice. Efficacy was carried with n=6-12/group.

Percent change in body weight was calculated as (BWcurrent - BWinitial)/(BWinitial) x 100%.

Data was presented as mean percent body weight change from the day of treatment initiation ± SEM. Tumor volume was determined as published previously (25). All data were expressed as mean ± standard error of the mean (SEM). Change in tumor volume were used for statistical analysis, and between group comparisons were carried out using a one-way ANOVA. For all statistical evaluations, the level of significance was set at p < 0.05. Significance compared to the vehicle control group is reported unless otherwise stated.

Drug formulation

LXH254 was and dosed p.o. in MEPC4 vehicle (45% Cremophor RH40 + 27% PEG400

+ 18% Corn Oil Glycerides (Maisine CC) + 10% ethanol) for all experiments except

PC3, where LXH254 was formulated in 90% PEG400 + 10% Tween80. MEPC4 stock was

diluted 5x (1:4 with DI Water) prior to dosing. Trametinib was dosed p.o. as a

suspension in 0.5% HPMC and 0.2% Tween80 in distilled water at pH8. Trametinib

was stirred overnight and protected from light at RT prior to dosing.

Xenograft qPCR

Tumors were homogenized and RNA was extracted using the Qiagen RNeasy Mini

Qiacube kit (Qiagen cat# 74116) per the manufacturer’s instructions. For RT-qPCR

reactions, 4ul/well of a 10ng/ul suspension of RNA loaded into 384-well plates along with 16l of Master Mix (Qiagen cat#204645), containing primers for hDUSP6 (Life

Tech cat# Hs00737962) and hRPLPO (Life Tech cat#4326314E), and reverse

transcriptase. Reactions were run for 20 min at 50°C, 15 min at 95° C, followed by 45

cycles of 45 s at 94°C and 45 s at 60°C.

Gene expression analysis

ARAF RNASeq expression, BRAF, NRAS, KRAS and HRAS mutation data from all

TCGA PanCancer Atlas studies were downloaded from the cBioPortal data viewer

(https://www.cbioportal.org/). RNASeq RSEM values were converted to TPM and log2

transformed. Only samples with ARAF expression were considered. Differential

expression normalization was performed using Limma/VOOM for differential expression

in R (30).

Structural Modeling

The in-house X-ray structure (PDB entry:6N0P) of LXH254 bound to BRAF was used in

conjunction with an X-ray structure of CRAF (PDB entry: 3OMV) to build a homology

model of ARAF with LXH254 bound in the molecular operating environment (MOE) (31).

This model was extensively refined with explicated solvent molecular dynamics simulations at 300K, 1atm, AM1BCC/ELF charges, TIP3P water, ff14SB (32) force field with the PARM@FROSST (33) extension within the AMBER (34) suite of simulation tools. 10ns simulations for each system were run and time average structures were extracted.

Results:

LXH254 inhibits both monomeric and dimeric RAF and promotes RAF dimer formation.

LXH254 is a potent inhibitor of the B-and CRAF kinases, inhibiting their catalytic activity at picomolar (pM) concentrations in biochemical assays (35). Consistent with LXH254 being a type II inhibitor, LXH254-bound BRAF adopts an inactive DFG out, C-helix in, conformation, and treatment of cells with LXH254 promotes BRAF/CRAF heterodimer formation most likely due to antagonism of the auto-inhibited state of the kinase which enables RAS-mediated dimer formation (35,36). We further extended characterization of RAF inhibition by LXH254 in several ways. First, we measured the biochemical

inhibition of ARAF, BRAF and CRAF by LXH254 (Fig. 1A). Consistent with previously

reported values (35), LHX254 inhibited BRAF and CRAF with IC50 values of 0.2 nM and

0.07 nM, respectively. Inhibition of ARAF, while robust, occurred with a higher IC50

value of 6.4 nM. Second, the ability of LXH254 to promote B/CRAF heterodimer formation of endogenous proteins was examined in three RAS-mutant cell lines via IP- western analysis (Fig. 1B). In all instances both LXH254, and the type II RAF inhibitor

LY3009120 (37), promoted heterodimer formation as judged by analysis of pull-downs using either BRAF or CRAF-directed antibodies. MEK1 and MEK2 were not detected in either BRAF or CRAF immuno-precipitates, however, both BRAF and CRAF were detected upon compound treatment following immunoprecipitation with MEK1/2 directed antibodies. Lastly, we compared the relative ability of LXH254 to inhibit monomeric

BRAFV600E in A-375 cells compared to RAS-induced wild-type RAF dimers in HCT 116

KRASG13D cells. For this assessment we used an approach described by Yao et al (22), in which cells are pre-incubated and subsequently washed following treatment with 1M encorafenib such that inhibition likely reflects LXH254 binding to the second protomer in

RAF dimers unoccupied by encorafenib. In contrast to the ~ 600-fold greater inhibition of monomeric BRAFV600 vs wt dimeric RAF by dabrafenib that we reported previously

V600 LXH254 displayed similar inhibition of monomeric BRAF and wt dimeric RAF (IC50 for p-ERK levels of 59 nM and 78 nM in A-375 and HCT 116 cells, respectively, Fig.

1C).

An issue that has potentially limited the efficacy of type II RAF inhibitors in clinic is a lack of selectivity for RAF versus other kinases (37-39). LXH254 displayed a high degree of selectivity for BRAF and RAF-1 (CRAF) relative to other kinases as judged by

KINOMEscan (35). We further interrogated the selectivity of LXH254 in two orthogonal screening formats. First, in biochemical assays using purified proteins LXH254 had IC50 values >10 M against 60/62 non-RAF kinases, with only MAPK14 (p38) and ABL1

having IC50 values below 10 M (2.1 M, and 4.1 M, respectively, Sup. Table 1).

Second, we profiled LXH254 using the KiNativ platform wherein kinase selectivity is

determined in cells via competition with an ATP-competitive covalent probe followed by

mass spectrometry (29). In HCT 116 cells incubated for 2 hours with 10 M LXH254,

the only kinases where probe binding was inhibited >80% were BRAF (90%) and CRAF

(88%) (Fig. 1D and Sup. Table 2). Beyond BRAF and CRAF, only two, ARAF (58%)

and p38 (66%), of the additional 184 kinases, detected in this cell line demonstrated

>50% inhibition of probe binding. High selectivity for BRAF and CRAF, and to a lesser

extent p38in HCT 116 cells was consistent with results from both the biochemical and

KINOMEscan profiling (35). Moreover, weaker binding of LXH254 to p38 in both

binding assay formats corresponds to relatively poor biochemical inhibition of this

kinase (IC50 = 2.1 M), relative to BRAF (0.0002 M) and CRAF (0.00007 M).

Collectively these data indicate that within the kinome, LXH254 is highly selective for

RAF kinases.

In vitro and in vivo activity of LXH254

We next profiled 291 cell lines for sensitivity to LXH254 in a high-throughput format

(Sup. Table 3). Using IC50 cut-offs ranging from 1-2.5 M, concentrations that approximate Cavg concentrations in clinic (data not shown), we found that cell lines harboring BRAFV600E/K as well as activating mutations in NRAS were enriched in the subset of sensitive cells. KRAS mutants were only weakly enriched, with cells lacking

RAS/RAF mutations being predominantly insensitive (22/146 with an IC50 < 2.0 M,

Sup. Fig 1A, Sup. Table 3). In the case of WT cell lines, in an insensitive model tested,

LXH254 effectively suppressed RAF signaling suggesting that some insensitivity may

result from a lack of pathway dependence (Sup. Fig 3F). To identify potential marker of

either resistance or sensitivity in MAPK-altered models we performed differential-gene

expression analysis, in both the entire set of cell lines as well as the KRAS, NRAS and

BRAF mutant subsets. Unfortunately, this analysis did not reveal any predictors of sensitivity beyond melanoma lineage makers in the BRAF subset, which was likely a

consequence of the high fraction of BRAFV600E models derived from this lineage (Sup.

Table 9). Cell line sensitivities to LXH254 were very similar to those previously obtained

for the type II RAF inhibitor RAF709 (Sup. Fig 1B (25)).

We subsequently examined the anti-tumor effects of LXH254 in a set of BRAF, NRAS and KRAS mutant xenograft models as well as a RAS/RAF wildtype model. In general, and concordant with cell line sensitivity in vitro, models harboring BRAF mutations

either alone (e.g. BRAFV600E) or co-incident with either activated NRAS (SK-MEL-30) or

KRAS (HEYA8, 1855HCOX) either regressed or demonstrated prolonged stasis

following treatment with LXH254 (Fig. 2 A-C, Sup. Table 7). In contrast to RAF-mutant

models, and similar to in vitro cell line data, the majority of KRAS mutant models, as

well as the RAS/RAF WT model displayed at best modest responses to LXH254

although there were notable outliers (e.g. Calu-6 cells, Fig. 2D, Sup. Table 7). When

examined in a subset of models, pathway inhibition was consistent with anti-tumor

results (Fig. 2 A-D).

Loss of ARAF expression sensitizes RAS-mutant cells to LXH254.

Based on the dual observations that (1) LXH254 had reduced binding of ARAF relative

to BRAF and CRAF in the Kinativ assay and, (2) that RAS mutants, particularly KRAS,

are largely insensitive to LXH254, we hypothesized that that insensitivity might stem

from poor ARAF inhibition by LXH254. To test this hypothesis we used CRISPR-Cas9

or zinc finger nucleases to functionally delete ARAF, BRAF, or CRAF from a series of

RAS mutant cell lines including two NRAS mutant melanoma cell lines (MEL-JUSO and

SK-MEL-30), and three KRAS mutant cell lines (COR-L23, MIA PaCa-2 and HCT 116,

Sup. Fig. 2). Loss of ARAF expression resulted in 3–11 fold increases in sensitivity to

LXH254 relative to parental cell lines in all models tested (Fig. 3, Sup. Table 3). Loss of

BRAF did not alter sensitivity to LXH254 in the wild-type BRAF cell lines; however, loss

of BRAF expression in NRAS mutant SK-MEL-30 cells which also harbor a kinase impaired class III variant (BRAFD287H,(40)) greatly reduced sensitivity to LXH254 (14-

fold). Loss of CRAF either did not change (MIA PaCa-2, COR-L23), or only slightly

altered (HCT 116) sensitivity to LXH254 in KRAS mutant models, however CRAF loss in

both NRAS mutant cell lines reduced sensitivity to LXH254 3-9 fold. In contrast, loss of

any of the three RAF paralogs did not consistently alter sensitivity to either the MEK1/2

inhibitor trametinib or ERK1/2 inhibitor ulixertinib (Sup. Tables 4 & 8).

We then examined the effect of loss of the various RAF proteins on pathway suppression by LXH254. Cells were treated with increasing concentrations of LXH254 for 4 hrs, and in some cases 24, 48 and 72 hrs, and p-ERK levels determined using the

Meso Scale Discovery platform (MSD), and/or western analysis using antibodies (Abs) that detect phosphorylated ERK1/2. Loss of ARAF improved pathway suppression by

LXH254 relative to parental cells and cells lacking either BRAF or CRAF (Table 1, Sup.

Fig. 3A). Similarly, reduced pathway inhibition occurred in cases where loss of BRAF

(SK-MEL-30) or CRAF (SK-MEL-30, MEL-JUSO) expression decreased sensitivity to

LXH254 (Table 1). Moreover, in both HCT 116 cells and MIA PaCa-2 cells where p-

ERK levels were monitored beyond 4 hours, the differential in pERK reduction by

LXH254 between cells lacking vs expressing ARAF increased further (Table 1, Sup. Fig.

3A). Thus, increased sensitivity to LXH254 in cells lacking ARAF corresponds to

improved pathway suppression. Reduced, albeit transient pathway inhibition by LXH254

in cells lacking CRAF suggests that in the absence of CRAF, ARAF might play a larger

role in signaling. To test this we immuno-precipitated ARAF and BRAF from parental

and CRAF-deficient MIA PaCa-2 cells following 2 and 24 hours of incubation with 1 M

LXH254. In parental cells increasing amounts of BRAF and CRAF co-purified with

immuno-precipitated ARAF over time after LXH254 treatment (Sup. Fig. 3B). Similarly,

increasing amounts of both ARAF, and more prominently CRAF, co-purified with BRAF.

In cells lacking CRAF expression, similar patterns of increasing levels of co-purified

BRAF following ARAF pull-down, and ARAF following BRAF pull-down were observed.

However, in the absence of CRAF, greater amounts of BRAF/ARAF heterodimers were

present at all times despite similar efficiencies of immuno-precipitation. Moreover, increased levels of BRAF/ARAF heterodimers in CRAF-deficient cells following the two- hour LXH254 treatment corresponded to failure of LXH254 to inhibit signaling. These data are consistent with a model where LXH254 has reduced ability to suppress MAPK signaling driven by ARAF and further that the contribution of ARAF to MAPK signaling

increases in the absence of CRAF expression.

ARAF-mediated resistance to LXH254 is dependent on both ARAF kinase activity,

and ability to form dimers.

To determine if ARAF-mediated LXH254 insensitivity depends on ARAF catalytic

function or the ability to form RAF-dimers, we re-introduced either N-terminally FLAG-

tagged wild type ARAF, kinase dead ARAF (K336M or D447A), or dimer-deficient ARAF

(R362H) variants into cells lacking endogenous ARAF. In all models examined, re- introduction of wild type ARAF reverted sensitivity to LXH254 to levels seen in parental cells (Figs. 4A & B). In contrast, expression of either kinase dead variants K336M (Fig.

4A&C) and D447A (Fig. 4B&C), or dimer-deficient variant ARAFR362H (Fig. 4B), did not

revert sensitivity to LXH254. Concordant with sensitivity to LXH254, over-expression of

wild type ARAF, but neither kinase dead nor dimer-deficient ARAF, reduced pathway

suppression by LXH254 (Sup. Table 5). Failure of kinase dead and dimer deficient

variants to complement loss of endogenous ARAF did not result from poor transgene

expression, as protein levels of all introduced ARAF variants exceeded the expression

of endogenous ARAF, although it should be noted that kinase dead variants were

consistently expressed at lower levels than either wild type or dimer-deficient versions

of ARAF (Sup. Fig. 2).

To ensure that kinase-dead and dimer-deficient RAF variants lacked kinase activity and

failed to form heterodimers, respectively, we performed IP-kinase assays using material

isolated from RAS mutant cell lines expressing the various FLAG-tagged ARAF

variants. In these experiments, the MEL-JUSO, HCT 116 and MIA PaCa-2, but not

COR-L23, cell lines lacked endogenous ARAF expression due to CRISPR/CAS9

editing. ARAF was immuno-precipitated using Abs directed against the FLAG epitope

from either untreated cells, or cells treated for 24 hours with LXH254 and RAF kinase

activity assessed using recombinant MEK1 as a substrate for the immuno-precipitated

material and monitoring production of phosphorylated MEK1 (p-MEK1) using a

phospho-MEK specific antibody. IP-kinase assays using precipitates from all ARAFwt expressing cells produced p-MEK1 (Fig 5A-D). In MEL-JUSO and COR-L23 cells IP-

kinase activity was largely dependent on pre-treatment of cells with LXH254 (Fig. 5A &

C), although examination of IP-kinase from lysates from wild type cells suggest that

stimulation of ARAF kinase activity by LXH254 pre-treatment occurred only in the

context of ARAF overexpression (Sup. Fig 3C). Phosphorylated MEK1 was unlikely to

result from auto-phosphorylation by recombinant MEK1, as IP-kinase assays carried out

in HEK293 cells using recombinant catalytically inert MEK1K97M also generated

phosphorylated MEK1 (Sup. Fig. 3D). Cells expressing ARAFK336M or ARAFD447A either

produced significantly less (COR-L23) or failed to generate detectable p-MEK1 despite

similar efficiencies of ARAF-immunoprecipitation. In all cell lines expressing exogenous

wild type and kinase-dead ARAF variants, treatment with LXH254 promoted

heterodimer formation between ARAF and both BRAF and CRAF, with the exception of

MEL-JUSO cells where only ARAF-CRAF heterodimers were detected. In contrast,

immuno-precipitates from HCT 116 and MEL-JUSO cells expressing ARAFR362H were

largely devoid of both B-and-CRAF and generated very little p-MEK1. These data

indicate that kinase dead and dimer-impaired variants lack catalytic activity and heterodimer forming capability, respectively. Failure of these variants to restore

LXH254 insensitivity indicates that both catalytic function and the ability to dimerize is

required for ARAF-mediated resistance to LXH254.

Cells expressing only ARAF are less sensitive to LXH254 than cells expressing

only BRAF or CRAF

To investigate further differential inhibition of RAF paralogs by LXH254, we generated

isogenic cell lines using CRISPR CAS-9 that expressed only one of the three RAF

genes. Clonal isolates expressing only ARAF (BRAF/CRAF or CRAF

(ARAF/BRAFwere generated in HCT 116, MIA PaCa-2 and MEL-JUSO cells.

BRAF-only cells were more difficult to generate, and while MIA PaCa-2

(ARAF/CRAF cells were generated, BRAF only HCT 116 cells maintained a low

level of CRAF (ARAF/CRAF-low) and we were unable to generate BRAF-only MEL-

JUSO cells (Sup. Fig. 2). Consistent with results in cells lacking ARAF (Fig. 4), BRAF- only and CRAF-only cells were 5-10 fold more sensitive to LXH254 in three-day growth

assays than parental cells (Fig 6A & Sup. Table 6). In contrast, ARAF-only cells were

less (HCT 116, MEL-JUSO) or similarly sensitive (MIA PaCa-2) to LXH254 than

parental cells (Fig. 6A). After four hours of treatment, LXH254 effectively inhibited

signaling in BRAF-and-CRAF-only cells, with IC50 values below 100 nM (Fig. 6B). In

contrast, in ARAF-only cells, signal inhibition required higher drug concentrations with

IC50 values measured at >1500 nM in all cell lines (Fig. 6B, Sup. Table 6). Moreover,

treatment of all ARAF-only cell lines with LXH254 resulted in modestly increased p-ERK

(25 – 50%) at LXH254 concentrations ranging from 10 to 370 nM in a manner reminiscent of paradoxical activation described for type 1/1.5 RAF inhibitors (41-43). As

described previously for cells lacking ARAF (Fig. 3 and Table 1) pathway re-activation

over time after LXH254 treatment in BRAF-and-CRAF-only cells was reduced relative to parental cells (Sup. Fig. 3E).

To compare the effects of LXH254 on signaling to that of a type 1.5 inhibitor we treated parental and individual RAF paralog-expressing MIA PaCa-2 cells for four hours with

the type 1.5 RAF inhibitor dabrafenib and measured the effects on p-ERK levels by

MSD. Consistent with previous results reported in RAS-mutant models, treatment of

MIA PaCa-2 cells with dabrafenib resulted in increased p-ERK levels to a peak ~ 4-fold

greater than baseline at concentrations up to 3.0 M (Fig. 6C). At concentrations

exceeding 3.0 M p-ERK levels declined, returning to original baseline levels at

dabrafenib concentrations of 10 M. Dabrafenib exerted distinctly different effects on p-

ERK levels in cells expressing only one of each of the three RAF proteins. In CRAF-

only cells, treatment with dabrafenib resulted in a similar pattern of pathway activation to

parental MIA PaCa-2cells, although activation was subtly greater at all concentrations

up to ~ 600 nM. In cells expressing only BRAF or ARAF, increases in p-ERK were also

observed however, increases were modest (~ 50%) relative to what occurred in parental

and CRAF-only cells. These data are consistent with previous reports suggesting that

CRAF activation is the major driver of paradoxical activation (17,41,44). Comparing the

effects of signaling indicated that while both drugs similarly and modestly induce

paradoxical activation in ARAF-only expressing cells, LXH254, but not dabrafenib is

able to inhibit wild type BRAF and CRAF in cells (Fig. 6B & C)

RAS mutants lacking ARAF are more sensitive to LXH254 in vivo.

We next determined the effect of ARAF ablation on sensitivity to MAPK inhibitors in

vivo. Tumors formed from parental and ARAF-deleted variants of the HCT 116, MIA

PaCa-2 and MEL-JUSO models were treated with vehicle, 0.3 mg/kg Q.D. trametinib, or

50 mg/kg B.I.D LXH254. Doses of trametinib and LXH254 were selected that matched

AUC values for the approved dose of trametinib (2mg Q.D.) or the recommended phase

2 dose (RP2D) of LXH254 (400mg B.I.D). ARAF deleted xenografts either grew with

similar kinetics (HCT 116 cells) or slightly slower (MIA PaCa-2, MEL-JUSO) than their

parental counterparts indicating that ARAF ablation has at most modest effects on the

fitness of these models. In all parental models, LXH254 and trametinib exerted similar

effects on tumor growth, resulting in a slow growth phenotype that was most

pronounced in MEL-JUSO cells (Fig. 7A-C). Treatment with LXH254 in the ARAF knockouts led to complete regression of HCT 116 and MEL-JUSO and near-complete regression of MIA PaCa-2 xenografts. Several tumors from each KRAS mutant model regrew after prolonged treatment, including 3/6 from the HCT 116, and 5/6 of the MIA

PaCa-2 tumors (Sup. Fig. 4 A-C). Initial analysis of these variants by western analysis

suggests that at least two tumors (MIA PaCa-2, M2 and M5) became resistant due to

RAF-pathway re-activation, with reactivation in one case potentially attributable to restoration of ARAF expression (M5). Lack of clear pathway re-activation in the other six tumors suggest resistance occurred via a MAPK-bypass mechanism (Sup. Fig. 4D).

Strikingly, 2/3 of the regressed HCT 116 tumors, failed to regrow after cessation of drug treatment consistent with complete eradication of the tumor (Sup. Fig. 4A). In contrast, variants lacking ARAF expression had similar sensitivities to trametinib as parental MIA

PaCa-2 cells.

These data indicate that RAS-driven activation of BRAF and CRAF is inhibited

sufficiently by LXH254 to eradicate tumors, and conversely that poor ARAF inhibition is

a critical contributor to the relative insensitivity of RAS-mutant models to LXH254.

Accordingly, we tested whether co-treatment of tumors with LXH254 and low doses of trametinib, aimed at inhibiting residual ARAF-driven signaling, improved anti-tumor effects in MIA-PaC-2 cells. Combining LXH254 with either 50% (0.3 mg/kg Q2D) or

10% (mg/kg QD) doses of trametinib-improved efficacy relative to both LXH254, and a

full dose of trametinib (0.3 mg/kg QD), although efficacy was inferior to that seen for

single agent LXH254 in cells lacking ARAF (Fig. 7D), and tumors regrew over time (data

not shown) . Thus, combining LXH254 with even small amounts of a MEK inhibitor can

result in significant anti-tumor effects.

Discussion

LXH254 is a type II RAF inhibitor with a high degree of selectivity for BRAF and CRAF,

that predominantly binds the inactive, DFG-out kinase configuration, and promotes the

formation of BRAF/CRAF heterodimers ((35) Fig 1, Sup. Tables 1 & 2). LXH254

treatment, also promoted MEK1/2 – B/CRAF interactions in IPs using MEK-directed

Abs, however, this interaction was not reciprocated in IPs using RAF-directed Abs. This

discrepancy may be an artifact due to the choice of RAF Abs used for the IPs.

Alternatively, this may reflect the existence of a substantial pool of dimerized RAF that

is not complexed with MEK1/2.

Here we describe further an unexpected paralog selectivity of this molecule wherein it is

a relatively poor inhibitor of ARAF. Poor ARAF-inhibitory activity in cells was somewhat

unanticipated based on biochemical analysis that indicated single digit nM inhibition of

the ARAF kinase in cell free assays. However, as has been described for CRAF

inhibition by type 1.5 inhibitors, in vitro inhibition of ARAF by LXH254 did not translate to

robust inhibition in the cellular setting. Ablation of ARAF expression in 5/5 RAS mutant

cell lines resulted in more potent anti-proliferative effects and RAF/MEK/ERK signal

suppression by LXH254 (Fig. 3, Table 1).

The mechanistic basis for the failure of LXH254 to inhibit ARAF is unknown and insights into a structural explanation is hampered by the lack of a crystal structure for ARAF.

However, as suggested here (Sup. Table 8) and elsewhere (37), failure to inhibit ARAF may not be a property of all type II RAF inhibitors. Simulations of ARAF structure using information available for BRAF and CRAF (see methods) suggest a possible explanation for reduced activity of LXH254 against ARAF (Sup. Fig. 5B). The terminal

hydroxyl group of the alkane chain of LXH254 forms a hydrogen bond with the

backbone carbonyl of F595 in BRAF (and CRAF) in the DFG-out configuration of the

protein (35). Simulations of the ARAF structure place this carbonyl shifted to a position

more proximal to the hydroxyl group on LXH254, which would likely eliminate this

hydrogen bond, instead creating a steric clash. Interestingly RAF709, a similarly poor

inhibitor of ARAF (Sup. Table 8) also forms a hydrogen bond with the same carbonyl on

F595, albeit via a water intermediary, which is also predicted to be lost in ARAF. An

understanding of the basis for the decreased inhibitory activity of LXH254 against ARAF

requires further study.

Roles for ARAF in promoting signal transduction in a scaffolding rather than catalytic

capacity have been previously described (45). However, we found that ARAF-

mediated insensitivity to LXH254 was dependent on ARAF having both kinase activity

and the ability to form dimers, strongly suggesting that failure of LXH254 to inhibit ARAF

catalytic activity while ensconced as an active member of RAF signaling complexes

underlies this phenotype. In RAS mutant tumors engineered to express only ARAF,

LXH254 paradoxically activated signaling to a similar degree, and with a similar bell-

shaped dose-response curve to the type 1.5-inhibitor dabrafenib (Fig. 6). In contrast, in

RAS-mutant cells expressing only B-or-CRAF LXH254 potently inhibited signaling

whereas dabrafenib caused pathway activation, most notably in cells expressing CRAF.

Collectively these data suggest a model where LXH254 potently inhibits dimeric BRAF and CRAF, but paradoxically activates ARAF-containing dimers (Sup. Fig. 5A).

In the phase I dose escalation trial of LXH254 very few partial responses to single agent

LXH254 were observed (46). This trial was enriched for KRAS-mutant tumors (~ 30%)

and consistent with preclinical data presented herein, the majority of patients with

KRAS-mutant disease progressed on treatment, although a minority of these patients displayed stabilization of disease. Based on findings here, LXH254 might have single agent activity only in RAS mutant tumors with either low expression or utilization of

ARAF. However, no correlation was found between ARAF mRNA expression and sensitivity to LXH254, either in the entire panel of cell lines profiled, or within specific

RAS/RAF-mutated subgroups in vitro (Sup. Fig. 1C). Moreover, analysis of RAS- mutant tumors in TCGA did not reveal a clearly identifiable subset of RAS-mutant tumors with intrinsically low-ARAF expression (Sup. Fig. 1D). Thus, the primary utility of

LXH254 is likely to be in combination with additional agents targeting the MAPK

pathway.

Consistent with this hypothesis, we found that adding even low doses of trametinib to a

full dose of LXH254 improved anti-tumor activity relative to single agent effects in the

MIA PaCa-2 model (Fig.7D). The observation that adult mice tolerate simultaneous loss

of B-and-CRAF (8) suggests that poor ARAF inhibition may improve LXH254 tolerability relative to e.g. MEK and ERK inhibitors, thereby facilitating the achievement of drug exposures where both BRAF and CRAF are strongly inhibited. Such improved tolerability might also facilitate combinations with additional MAPK inhibitors for the treatment of RAS-mutant tumors resulting in improved pathway suppression and anti- tumor effects. In support of this idea, LXH254-anchored MAPK combinations, such as

those described in Fig. 7, display a superior tolerability/efficacy relationship when

compared to single agent treatments in preclinical species (Manuscript in prep).

Moreover, combinations between LXH254 and the MEK inhibitor trametinib and ERK

inhibitor LTT462 are currently in dose expansion in patients with a variety of RAS/RAF-

pathway alterations.

Acknowledgements

We thank David Kodack, Qing Shen, Tina Yuan, Rita Das, Flavio Somanji, Huyen

Nguyen, Huaixiang Hao and Albot Liu for critical reading and comments in the

manuscript. We thank Jodi Meltzer, Qamrul Hassan, Yingyun Wang, and Malika Singh

for technical contributions to the generation of RAF-paralog knockout models, BRAF

and CRAF biochemical assays, and in vivo PC3 experiment.

References

1. Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer discovery 2012;2(5):401-4 doi 10.1158/2159-8290.CD-12-0095. 2. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Science signaling 2013;6(269):pl1 doi 10.1126/scisignal.2004088. 3. Janes MR, Zhang J, Li LS, Hansen R, Peters U, Guo X, et al. Targeting KRAS Mutant Cancers with a Covalent G12C-Specific Inhibitor. Cell 2018;172(3):578-89 e17 doi 10.1016/j.cell.2018.01.006. 4. Lito P, Solomon M, Li LS, Hansen R, Rosen N. Allele-specific inhibitors inactivate mutant KRAS G12C by a trapping mechanism. Science 2016;351(6273):604-8 doi 10.1126/science.aad6204. 5. Ostrem JM, Peters U, Sos ML, Wells JA, Shokat KM. K-Ras(G12C) inhibitors allosterically control GTP affinity and effector interactions. Nature 2013;503(7477):548-51 doi 10.1038/nature12796. 6. 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 discovery 2016;6(3):316-29 doi 10.1158/2159-8290.CD-15-1105. 7. Uehling DE, Harris PA. Recent progress on MAP kinase pathway inhibitors. Bioorg Med Chem Lett 2015;25(19):4047-56 doi 10.1016/j.bmcl.2015.07.093. 8. 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(5):652-63 doi 10.1016/j.ccr.2011.04.002. 9. Abe H, Kikuchi S, Hayakawa K, Iida T, Nagahashi N, Maeda K, et al. Discovery of a Highly Potent and Selective MEK Inhibitor: GSK1120212 (JTP-74057 DMSO Solvate). ACS Med Chem Lett 2011;2(4):320-4 doi 10.1021/ml200004g. 10. Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Larkin J, et al. Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med 2011;364(26):2507-16 doi 10.1056/NEJMoa1103782. 11. Davies BR, Logie A, McKay JS, Martin P, Steele S, Jenkins R, et al. AZD6244 (ARRY-142886), a potent inhibitor of mitogen-activated protein kinase/extracellular signal-regulated kinase kinase 1/2 kinases: mechanism of action in vivo, pharmacokinetic/pharmacodynamic relationship, and potential for combination in preclinical models. Mol Cancer Ther 2007;6(8):2209-19. 12. 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(18):5339-48 doi 10.1158/1078-0432.CCR-16-2923. 13. Dummer R, Ascierto PA, Gogas HJ, Arance A, Mandala M, Liszkay G, et al. Encorafenib plus binimetinib versus vemurafenib or encorafenib in patients with BRAF-mutant melanoma (COLUMBUS): a multicentre, open-label, randomised phase 3 trial. Lancet Oncol 2018;19(5):603- 15 doi 10.1016/S1470-2045(18)30142-6.

14. Flaherty KT, Robert C, Hersey P, Nathan P, Garbe C, Milhem M, et al. Improved survival with MEK inhibition in BRAF-mutated melanoma. N Engl J Med 2012;367(2):107-14 doi 10.1056/NEJMoa1203421. 15. Gilmartin AG, Bleam MR, Groy A, Moss KG, Minthorn EA, Kulkarni SG, 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 2011;17(5):989-1000 doi 10.1158/1078- 0432.CCR-10-2200. 16. Hauschild A, Grob JJ, Demidov LV, Jouary T, Gutzmer R, Millward M, et al. Dabrafenib in BRAF- mutated metastatic melanoma: a multicentre, open-label, phase 3 randomised controlled trial. Lancet 2012;380(9839):358-65 doi 10.1016/S0140-6736(12)60868-X. 17. Rheault TR, Stellwagen JC, Adjabeng GM, Hornberger KR, Petrov KG, Waterson AG, et al. Discovery of Dabrafenib: A Selective Inhibitor of Raf Kinases with Antitumor Activity against B- Raf-Driven Tumors. ACS Med Chem Lett 2013;4(3):358-62 doi 10.1021/ml4000063. 18. Yang H, Higgins B, Kolinsky K, Packman K, Go Z, Iyer R, et al. RG7204 (PLX4032), a selective BRAFV600E inhibitor, displays potent antitumor activity in preclinical melanoma models. Cancer Res 2010;70(13):5518-27 doi 10.1158/0008-5472.CAN-10-0646. 19. Yeh TC, Marsh V, Bernat BA, Ballard J, Colwell H, Evans RJ, et al. Biological characterization of ARRY-142886 (AZD6244), a potent, highly selective mitogen-activated protein kinase kinase 1/2 inhibitor. Clin Cancer Res 2007;13(5):1576-83 doi 10.1158/1078-0432.CCR-06-1150. 20. Blumenschein GR, Jr., Smit EF, Planchard D, Kim DW, Cadranel J, De Pas T, et al. A randomized phase II study of the MEK1/MEK2 inhibitor trametinib (GSK1120212) compared with docetaxel in KRAS-mutant advanced non-small-cell lung cancer (NSCLC)dagger. Ann Oncol 2015;26(5):894- 901 doi 10.1093/annonc/mdv072. 21. Simanshu DK, Nissley DV, McCormick F. RAS Proteins and Their Regulators in Human Disease. Cell 2017;170(1):17-33 doi 10.1016/j.cell.2017.06.009. 22. 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(3):370-83 doi 10.1016/j.ccell.2015.08.001. 23. Lake D, Correa SA, Muller J. Negative feedback regulation of the ERK1/2 MAPK pathway. Cell Mol Life Sci 2016;73(23):4397-413 doi 10.1007/s00018-016-2297-8. 24. 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(7391):603-7 doi 10.1038/nature11003. 25. Shao W, Mishina YM, Feng Y, Caponigro G, Cooke VG, Rivera S, et al. Antitumor Properties of RAF709, a Highly Selective and Potent Inhibitor of RAF Kinase Dimers, in Tumors Driven by Mutant RAS or BRAF. Cancer Res 2018;78(6):1537-48 doi 10.1158/0008-5472.CAN-17-2033. 26. Wylie AA, Schoepfer J, Jahnke W, Cowan-Jacob SW, Loo A, Furet P, et al. The allosteric inhibitor ABL001 enables dual targeting of BCR-ABL1. Nature 2017;543(7647):733-7 doi 10.1038/nature21702. 27. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, et al. Multiplex genome engineering using CRISPR/Cas systems. Science 2013;339(6121):819-23 doi 10.1126/science.1231143. 28. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, et al. RNA-guided human genome engineering via Cas9. Science 2013;339(6121):823-6 doi 10.1126/science.1232033. 29. Nomanbhoy TK, Sharma G, Brown H, Wu J, Aban A, Vogeti S, et al. Chemoproteomic Evaluation of Target Engagement by the Cyclin-Dependent Kinase 4 and 6 Inhibitor Palbociclib Correlates with Cancer Cell Response. Biochemistry 2016;55(38):5434-41 doi 10.1021/acs.biochem.6b00629.

30. Law CW, Chen Y, Shi W, Smyth GK. voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol 2014;15(2):R29 doi 10.1186/gb-2014-15-2-r29. 31. Group CC. 2018 MOE 2018.01. 32. Maier JA, Martinez C, Kasavajhala K, Wickstrom L, Hauser KE, Simmerling C. ff14SB: Improving the Accuracy of Protein Side Chain and Backbone Parameters from ff99SB. J Chem Theory Comput 2015;11(8):3696-713 doi 10.1021/acs.jctc.5b00255. 33. Christopher Bayly DMaJ-FT. 2010 An Informal AMBER Small Molecule Force Field: parm@Frosst. 34. D.A. Case RMB, D.S. Cerutti, T.E. Cheatham, III, T.A. Darden, R.E. Duke, T.J. Giese, H. Gohlke, A.W. Goetz, N. Homeyer, S. Izadi, P. Janowski, J. Kaus, A. Kovalenko, T.S. Lee, S. LeGrand, P. Li, C. Lin, T. Luchko, R. Luo, B. Madej, D. Mermelstein, K.M. Merz, G. Monard, H. Nguyen, H.T. Nguyen, I. Omelyan, A. Onufriev, D.R. Roe, A. Roitberg, C. Sagui, C.L. Simmerling, W.M. Botello-Smith, J. Swails, R.C. Walker, J. Wang, R.M. Wolf, X. Wu, L. Xiao and P.A. Kollman. AMBER 2016 Reference Manual., University of California, San Francisco. 2016. 35. Ramurthy S, Taft BR, Aversa RJ, Barsanti PA, Burger MT, Lou Y, et al. Design and Discovery of N- (3-(2-(2-Hydroxyethoxy)-6-morpholinopyridin-4-yl)-4-methylphenyl)-2-(trifluorom ethyl)isonicotinamide, a Selective, Efficacious, and Well-Tolerated RAF Inhibitor Targeting RAS Mutant Cancers: The Path to the Clinic. J Med Chem 2019 doi 10.1021/acs.jmedchem.9b00161. 36. Jin T, Lavoie H, Sahmi M, David M, Hilt C, Hammell A, et al. RAF inhibitors promote RAS-RAF interaction by allosterically disrupting RAF autoinhibition. Nature communications 2017;8(1):1211 doi 10.1038/s41467-017-01274-0. 37. 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(3):384-98 doi 10.1016/j.ccell.2015.08.002. 38. Wilhelm SM, Carter C, Tang L, Wilkie D, McNabola A, Rong H, et al. BAY 43-9006 exhibits broad spectrum oral antitumor activity and targets the RAF/MEK/ERK pathway and receptor tyrosine kinases involved in tumor progression and angiogenesis. Cancer Res 2004;64(19):7099-109 doi 10.1158/0008-5472.CAN-04-1443. 39. Williams TE, Subramanian S, Verhagen J, McBride CM, Costales A, Sung L, et al. Discovery of RAF265: A Potent mut-B-RAF Inhibitor for the Treatment of Metastatic Melanoma. ACS Med Chem Lett 2015;6(9):961-5 doi 10.1021/ml500526p. 40. Yao Z, Yaeger R, Rodrik-Outmezguine VS, Tao A, Torres NM, Chang MT, et al. Tumours with class 3 BRAF mutants are sensitive to the inhibition of activated RAS. Nature 2017;548(7666):234-8 doi 10.1038/nature23291. 41. 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(7287):431-5 doi 10.1038/nature08833. 42. 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(2):209-21 doi 10.1016/j.cell.2009.12.040. 43. Poulikakos PI, Zhang C, Bollag G, Shokat KM, Rosen N. RAF inhibitors transactivate RAF dimers and ERK signalling in cells with wild-type BRAF. Nature 2010;464(7287):427-30 doi 10.1038/nature08902. 44. Lito P, Pratilas CA, Joseph EW, Tadi M, Halilovic E, Zubrowski M, et al. Relief of profound feedback inhibition of mitogenic signaling by RAF inhibitors attenuates their activity in BRAFV600E melanomas. Cancer Cell 2012;22(5):668-82 doi 10.1016/j.ccr.2012.10.009. 45. Rebocho AP, Marais R. ARAF acts as a scaffold to stabilize BRAF:CRAF heterodimers. Oncogene 2013;32(26):3207-12 doi 10.1038/onc.2012.330.

46. Janku F, Iyer G, Spreafico A, Yamamoto N, Bang Y-J, Elez E, et al. A phase I study of LXH254 in patients (pts) with advanced solid tumors harboring MAPK pathway alterations. Journal of Clinical Oncology 2018;36(15_suppl):2586- doi 10.1200/JCO.2018.36.15_suppl.2586. 47. Cancer Genome Atlas Research N, Weinstein JN, Collisson EA, Mills GB, Shaw KR, Ozenberger BA, et al. The Cancer Genome Atlas Pan-Cancer analysis project. Nat Genet 2013;45(10):1113-20 doi 10.1038/ng.2764.

Table 1:

pERK IC50 (nM) Cell Line Variant SK-MEL-30 MEL-JUSO HCT 116 COR-L23 Mia PaCa-2 Mia PaCa-2 (24 hrs) Wild Type 90 77 12 889 687 >10000 ARAF 60 10 5 43 85 170 BRAF 650 94 18 1072 262 >10000 CRAF 300 531 357 1358 3627 >10000

Figure Legends

Figure1. Characterization of LXH254 in biochemical and cellular assays. (A) Shown are the calculated IC50 values against purified ARAF, BRAF and CRAF in biochemical assays and structure of LXH254. (B) Western analysis of whole cell lysates (WCL) and

BRAF, CRAF and MEK immuno-precipitates (IP), from HCT 116, MEL-JUSO and MIA

PaCa-2 cells either untreated, or following 1 hr. treatments with either 1M LXH254 or

1M LY3009120. Proteins detected by immunoblotting are indicated to the right of each panel. (C) Western analysis (top panels) and quantitation via densitometry (lower panel) of the indicated proteins in A375 and HCT 116 cells following treatment with increasing amounts of LXH254. LXH254 concentrations (M) are indicated above each lane and immunoblotted proteins to the right of each panel. HCT 116 cells were treated for 1hr

with encorafenib that was subsequently washed-out prior to treatment with LXH254. (D)

Assessment of interactions between LXH254 and kinases in HCT 116 cells. Individual kinases are depicted as circles along the X-axis and fractional inhibition of kinase interactions with an ATP-competitive probe by LXH254 are given on the Y-axis.

Figure 2. Representative xenograft models treated with LXH254. Shown on the left of each panel are changes in tumor volume over time (days) and on the right of each panel changes in DUSP6 mRNA level after a single administration of LXH254 for the

WM-793 (BRAFV600E, A), SK-MEL-30 (NRASQ61K, BRAFD287H, B), HEYA8 (KRASG12D,

BRAFG464E, C) and 3390HPAX (KRASG12D, D) models. Vehicle (Veh) treated animals are indicated with ( ) and LXH254 treated animals with ( ). All models were treated with LXH254 100 mg/kg qd. T/C determinations were made at the time when control animal were sacrificed due to tumor size (1000 - 1500 mm3) and denoted with an asterix.

Figure 3. Loss of ARAF expression sensitizes RAS-mutant cell lines to LXH254.

Shown are proliferation curves for five cell lines ( ) and derived variants lacking expression of ARAF ( ), BRAF ( ) and CRAF ( ). Cell line identity and RAS (or RAF) mutation status are shown above each graph. Concentrations of LXH254 (nM) and percentage of cells remaining relative to an untreated control as judged by ATP levels are given on the X, and Y-axis, respectively. Effects of LXH254 on proliferation were

determined after either three (MEL-JUSO, SK-MEL-30, HCT 116, MIA PaCa-2) or five

(COR-L23) days of incubation with LXH254.

Figure 4. Expression of either kinase-dead or dimer-deficient ARAF fails to reverse sensitivity to LXH254 in cells lacking ARAF expression. Shown are proliferation curves for the COR-L23, MIA PaCa-2, HCT 116 and MEL-JUSO cell lines, both wild type ( ) and variants lacking ARAF expression ( ). Also shown are proliferation curves for cell lines lacking endogenous ARAF expression, but expressing exogenous wild type ARAF

( ), kinase impaired ARAF ( ), either K336M (A), or D447A (B) or dimer-deficient

R362H ARAF ( ). Effects of LXH254 on proliferation were determined after either three

(MEL-JUSO, SK-MEL-30, HCT 116, MIA PaCa-2) or five (COR-L23) days of incubation with LXH254. In the case of MIA PaCa-2 cells shown in (A) expression of exogenous

ARAF variants were under the transcriptional control of the doxycycline (dox) –regulated

TET-3G promoter with cells grown in the absence of dox depicted with hashed lines.

IC50 values determined for the curves shown in A&B are provided in (C). Values denoted with * and ** were determined in the absence of presence of doxycycline, respectively.

Figure 5. Kinase-impaired and dimer deficient ARAF variants lack in vitro kinase activity and have reduced interactions with B-and-CRAF, respectively. Shown are western blots using the indicated antibodies performed on either whole cell lysates

(WCL) or material immuno-precipitated using a FLAG-directed Ab (IP:FLAG) from COR-

L23 (A), MIA PaCa-2 (B), MEL-JUSO (C) and HCT 116 (D) cell lines. In the cases of

COR-L23 and MIA PaCa-2 cells exogenously expressed ARAF variants were under the transcriptional control of the doxycycline (dox) –regulated TET-3G promoter.

Concentrations of LXH254 used for cell line pretreatment are shown above each sample. For the IP-kinase assay purified recombinant HISx6 -tagged MEK1 and ATP were added to immuno-precipitated material.

Figure 6. RAS-mutant cells expressing only ARAF are resistant to LXH254. Shown are proliferation (A) and p-ERK (B) inhibition curves for parental ( ) HCT 116, MIA

PaCa-2 and MEL-JUSO cell lines as well as variants lacking B-and-CRAF expression (

), ARAF and BRAF expression ( ) and A-and-CRAF expression ( ) following treatment with increasing concentrations of LXH254. In all cases, changes in cell number were determined after 72 hours and p-ERK levels after four hours of incubation with LXH254. p-ERK levels following four hours of treatment of MIA PaCa-2 cells with increasing concentrations of either LXH254 or dabrafenib are shown in (C).

Figure 7. RAS-mutant cell lines lacking ARAF are sensitive to LXH254 in vivo. Shown are changes in tumor volume until the time of T/C determination results for HCT 116

(A), MIA PaCa-2 (B, D) and MEL-JUSO (C) models. Parental model are denoted with

(solid lines) and variants lacking ARAF expression with dashed lines. Vehicle treated animals are shown in black ( ), 0.3 mg/kg trametinib Q.D. treated in blue ( ) and 50 mg/kg LXH254 B.I.D in red ( ). . Shown in (D) is the MIA PaCa-2 model treated as in

(B) but with additional combinations of 50 mg/kg LXH254 BID with either 0.3 mg/kg

Q2D ( ) or 0.03 mg/kg ( ) trametinib.

Figure 1

A Assay IC50 (nM) B

ARAF 6.4

DMSO DMSO DMSO

LXH254 LXH254 LXH254

LY3009120 LY3009120 LY3009120 BRAF 0.21 pMEK (S217/221) CRAF 0.072 MEK pERK (T202/Y204)

IERK WCL

BRAF

CRAF

GAPDH

BRAF

CRAF BRAF IP BRAF MEK

BRAF

CRAF CRAF IP CRAF MEK

BRAF

CRAF MEK IP MEK MEK C D HCT 116 MEL-JUSO Mia PaCa-2

BRAF CRAF p38α ARAF

1 2 5 A 3 7 5

1 0 0 H C T 1 1 6

K R

% Inhibition E

p 7 5

v i

t 5 0

e R 2 5

0 - 9 - 8 - 7 - 6 - 5

L o g [ M ] Kinase Downloaded from clincancerres.aacrjournals.org on October 6, 2021. © 2020 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 22, 2020; DOI: 10.1158/1078-0432.CCR-20-2563 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Figure 2

A B

C D

Figure 3

M E L -J U S O (N R A S Q 6 1 L ) S K -M E L -3 0 (N R A S Q 6 1 K , B R A F D 2 8 7 H )

1 2 5 1 2 5

1 0 0 1 0 0

h h

t t

w w

o o r

7 5 r 7 5 Parental

g g

l l

o o r

r ARAF Knockout

t t n

5 0 n 5 0 o

o BRAF Knockout

C C

% % CRAF Knockout 2 5 2 5

0 0 1 1 0 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 1 0 0 1 0 0 0 1 0 0 0 0 L X H 2 5 4 (n M ) L X H 2 5 4 (n M )

C O R -L 2 3 (K R A S G 1 2 V ) H C T 1 1 6 (K R A S G 1 3 D ) M IA P a C a -2 (K R A S G 1 2 C )

1 2 5 1 2 5 1 2 5

1 0 0 1 0 0 1 0 0

h h h

t t t

w w w

o o o

r r

7 5 7 5 r 7 5

g g g

l l l

o o o

r r r

t t t

n n

5 0 5 0 n 5 0

o o o

C C C

% % % 2 5 2 5 2 5

0 0 0 1 1 0 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 1 0 0 1 0 0 0 1 0 0 0 0 L X H 2 5 4 (n M ) L X H 2 5 4 (n M ) L X H 2 5 4 (n M )

Figure 4

G 1 2 V C OCRO-LR2-3L (2K3R A(KSRG 1A2 VS) ) A. COR-L23 (KRASG12V) MIA PaCa-2 (KRASG12C) 1 2255 125 P a re n ta l Parental P a re n t A R A F K n o c k o u t ARAF Knockout + 1 0 0 A R A F K n o c k o u t h ARAF-WT-FLAG (- Dox)

t 100

1 0 0 A R A F K n o c k o u t +

t t

o A R A F -W T -AF LRAAGF K n o c k o u t +

r w

w 7 5 ARAF Knockout +

o l A R A F -W T -F L A G r r ARAF-WT-FLAG (+ Dox) o A R A F K n o c k o u t + 75

7 5 g

l A R A F -K 3 6 6 M -F L A G ARAF Knockout + n

A R A F K n o c k o u t +o o

o 5 0 r

r ARAF-K366M-FLAG (- Dox)

A R A F -K 3 6 6 M -F Ln A G

n %

5 0 o o 50

ARAF Knockout + C

C 2 5

ARAF-K366M_FLAG (+ Dox)

% %

205 25 1 1 0 1 0 0 1 0 0 0 1 0 0 0 0 L X H 2 5 4 (n M ) 0 H C T 1 1 6 0 1 1 0 1 0 0 1 0 0 0 1 0 0 0 0 1 10 100 1000 10000 H C T 1 1 6 A K O L X H 2 5 4 (n M ) LXH254 (nM) MELJUSO B. H C T 1 1 6 A K O + O E w t A R A F L X H 2 5 4 A c t i v i t y M I A P a C a 2 MELJUSO ARAF KO M I A A R A F k / o HCT 116 (KRASG13D) MEL-JUSOH C T 1 1 6 A (NRASK O + O EQ61L d d )A R A F 1 2 5 MIA PaCa-2 (KRASG12C) +ARAF-WT P a re n ta l 1 2 5 H C T 1 1 6 A K O + O E k d A R A F 1 2 5 M I A O E A R A F 125 A R A F K n o c k o u t 1 0 0

h M I A O E k d A R A F

+t ARAF-D447A KD A R A F K n o c k o u t + w

1 0 0 1o 0 0 A R A F -W T -F L A G

100 r 7 5 M I A O E d d A R A F

t l

h A R A F K n o c k o u t + t

+o ARAF-R362H DD

r t

w A R A F -D 4 4 7 A -F L A G

r r o 5 0 o 7 5

7 5 r g

g 75

g A R A F K n o c k o u t +

o o

% A R A F -R 3 6 2 H -F L A G

r t

t 2 5

n n

n 5 0

5 0 o o

o 50

% % % 1 1 0 1 0 0 1 0 0 0 1 0 0 0 0 2 5 25 2 5 L X H 2 5 4 (n M )

0 0 0 1 0 1 0 0 1 0 0 0 1 0 0 0 0 1 10 100 1000 10000 1 0 1 0 0 1 0 0 0 1 0 0 0 0 L X H 2 5 4 ( n M ) LXH254 (nM) L X H 2 5 4 ( n M )

Proliferation IC50 (nM) C. Cell Line Variant MEL-JUSO HCT 116 MIA PaCa-2 MIA PaCa-2 (inducible) COR-L23 Wild Type 1015 1796 1560 1278 4274 ARAFD 372 491 565 707* 1349 ARAFD + ARAFWT 2420 1794 1854 1427** 6668 ARAFD + ARAFK362H 442 201 560 N/A N/A ARAFD + ARAFA447D 453 269 605 N/A N/A ARAFD + ARAFK336M N/A N/A N/A 710 1675 Downloaded from clincancerres.aacrjournals.org on October 6, 2021. © 2020 American Association for Cancer Research. Author Manuscript Published OnlineFirst on December 22, 2020; DOI: 10.1158/1078-0432.CCR-20-2563 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Figure 5

A. B. C.

P a r e n t a l

A R A F o n ly Figure 6 A. B. B R A F o n ly C. C R A F o n ly M I A P a C a 2 G13D G13D G12C HCT 116 (KRAS ) HCT 116HCT (KRAS 116 ) MIAMIA PaCa PaCa--22 (KRAS ) M I A P a C a 2 A R A F o n ly 1 2 5 1 7 5 H C T 1 1 6 6 0 0 M I A P a C a 2 B R A F o n ly 1 5 0 A R A F o n ly

1 0 0 2

/ M I A P a C a 2 C R A F o n ly

1 2 t B R A F o n ly

1 2 5 /

K w

1 P a r e n t a l

R K

r 7 5 C R A F o n ly 4 0 0

g R

1 0 0

E l

o A R A F o n ly

l n t 7 5

5 0 o

t o

C B R A F o n ly

5 0 o

% 2 0 0

c % 2 5 C R A F o n ly 2 5 %

0 0 1 0 1 0 0 1 0 0 0 1 0 0 0 0 7 2 5 4 1 3 0 . 0 ...... 0 0 3 1 3 0 1 3 0 1 2 5 1 4 2 7 1 3 0 L X H 2 5 4 ( n M ) 1 3 1 3 0 0 4 8 1 3 0 0 0 1 3 0 ...... 1 0 2 9 9 6 5 0 M I A P0 a C a 2 3 5 2 0 0 1 6 5 0 2 0 M I A P La CX aH 22 5 4 ( n M ) 1 1 0 0 M I A P a C a 2 B R A F o n ly

D a b r a f e n i b ( n M ) h G12C M I A A R A F o n ly G12Ct 1 2 5 MIA PaCa-2 (KRAS ) 1 7 5 MIA MPaCa-I AMIA C R A F PaCa-2o2n ly(KRAS w )

o M I A P a C a 2

r M I A P a C a 2 L X H 2 5 4 7 5

M I A B R A F o n ly g

1 5 0 M I A P a C a 2 A R A F o n ly 1 0 0 l 2 2 0 0 D a b r a f e n ib

/ 6 0 0

r t 1 M I A P a C a 2 B R A F o n ly M I A P a C a 2 B R A F o n ly

1 2 5 t

o R r 5 0 M I A P a C a 2 C R A F o n ly

7 5 o

g E

6 0 0 2

1 0 0 / l

/ 1 5 0

t K

o 4 0 0

t R

5 0 7 5 %

o p

2 5

l 1 0 0

o /

5 0 o

1 %

2 5 n

o K o 2 0 40 0 0

2 5 c

R 5 0

0 % % 0 E

0 p

1 0 1 0 0 1 0 0 0 1 0 0 0 0

1 0 1 0 0 1 0 0 0 1 0 0 0 0 l 7 2 5 4 1 3 0 . 0 ...... M E L J U S O 0 3 1 3 0 1 3 0 o

r M I A P a C a 2 A R A F o n ly 1 4 2 7 1 3 0

L X H 2 5 4 ( n M ) t 0 1 3 1 3 0 0 L X H 2 5 4 ( n M ) 1 3 0 1 n M E L J U S O A R A F o n ly 0 7 2 5 4 1 3 0 . 0 . 7 . 2 . 5 . 4 . 1 . 3 . 0

o ...... 2 0 0 0 3 1 3 0 0 0 3 1 3 0 1 3 0 1 3 1 4 2 7 1 3 0

c 1 4 2 7 0 1 3 1 3 1 3 0 L X H 2 5 4 ( n M ) 1 3 1 3 0 1 3 0 1 3 0 M I A P a C a 2 C R A F o n ly

M E L J U S O C R A F o n ly 1 1 %

1 2 5 C o m p o u n d ( n M ) C o m p o u n d ( n M ) Q61L 1 7 51 7 5 Q61L MEL-JUSO (NRAS ) M I A P a C a 2 A R A F o n ly MEL-JUSO (NRAS ) MEL-JUSO 2 0 0 6 0 0

1 0 0 1 5 01 5 0 2

2 0

1 t

1 M I A P a C a 2 C R A F o n ly

2 2

K 1 2 5 w

K 1 2 5 0 7 2 5 4 1 3 0 /

/ 1 5 0 ......

1 1

R 0 3 1 3 0 1 3 0

r 7 5 1 4 2 7 1 3 0

E g E 4 0 0

1 0 0 1 3 1 3 0 l

p 1 0 0

p R

R 1 3 0

l o

l 1

l l

5 0 t 7 5 7 5 1 0 0

t t

o C o m p o u n d ( n M )

5 0 o

5 0 o 2 0 0

2 5 5 0 % 2 5 2 5 %

0 0 0 0 0 0 4 8 1 1 3 3 0 0 0 0 0 0 1 0 1 0 0 1 0 0 0 1 0 0 0 0 . 0 . . 4 . . 8 ...... 0 0 2 2 9 9 9 9 6 6 5 5 0 0 0 0 3 3 5 5 2 2 0 0 0 0 0 7 2 5 4 1 3 0 0 7 2 5 4 1 3 0 1 6 5 0 0 ...... 1 6 5 0 3 1 3 0 1 3 0 0 3 1 3 0 1 3 0 L X H 2 5 4 ( n M ) 2 2 0 0 1 1 4 2 7 1 3 0 1 4 2 7 1 3 0 1 1 3 1 3 0 1 3 1 3 0 1 3 0 1 3 0 1 1 L X HL X2 H5 42 5( n4 M ( n) M ) Downloaded from clincancerres.aacrjournals.org on October 6, 2021. © 2020 American AssociationC o m p oforu n dCancer ( n M ) C o m p o u n d ( n M ) Research. Author Manuscript Published OnlineFirst on December 22, 2020; DOI: 10.1158/1078-0432.CCR-20-2563 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Figure 7

A. B. C.

LXH254, a potent and selective ARAF-sparing inhibitor of BRAF and CRAF for the treatment of MAPK-driven tumors

Kelli-Ann Monaco, Scott Delach, Jing Yuan, et al.

Clin Cancer Res Published OnlineFirst December 22, 2020.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-20-2563

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2020/12/22/1078-0432.CCR-20-2563.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet been Manuscript edited.

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2020/12/22/1078-0432.CCR-20-2563. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.