RAF inhibitor resistance is mediated by dimerization of aberrantly spliced BRAF(V600E)

Tumors with mutant BRAF(V600E) are dependent on BRAF/MEK/ERK (ERK signaling) for their growth. ATP-competitive RAF inhibitors suppress ERK signaling in cells with mutant BRAF, but paradoxically enhance ERK signaling in cells with wild-type BRAF. We have shown previously the mechanistic basis of this phenomenon: active RAS promotes homo- and heterodimerization of RAF. When RAF exists in dimers, drug binding to one protomer transactivates the other unbound protomer. In BRAF(V600E) tumors, RAS is not activated, RAF is predominantly monomeric and ERK signaling is inhibited by the drug. These results predict that RAF inhibitors will be effective in tumors in which BRAF is mutated. Furthermore, because RAF inhibitors do not inhibit ERK signaling in other cells, the model predicts that they would have a higher therapeutic index and greater antitumor activity than MEK inhibitors. Confirming these predictions, RAF inhibitors have shown remarkable clinical activity in patients with melanomas that harbor mutant BRAF(V600E), however, resistance invariably develops. The mechanism of action of these drugs indicates that promotion of RAF dimerization (by lesions that activate RAS, or other means) could lead to drug resistance in mutant BRAF tumors. Indeed, RAS mutation has been reported as a mechanism of resistance in a subset of patients that progressed on the RAF inhibitor PLX4032 (vemurafenib). In order to identify novel mechanisms of resistance, we generated cell lines resistant to PLX4032. We found that a subset of cells resistant to PLX4032 express a 61kd splicing variant form of BRAF(V600E) that lacks exons 4–8, a region that encompasses the RAS-binding domain (p61BRAF(V600E)). p61BRAF(V600E) exhibits enhanced dimerization as compared to full length BRAF(V600E) in cells with low levels of RAS activation. In cells in which p61BRAF(V600E) is expressed endogenously or ectopically, ERK signaling is resistant to the RAF inhibitor. Moreover, a mutation that abolishes the dimerization of p61BRAF(V600E) restores its sensitivity to PLX4032, indicating that dimerization status of BRAF(V600E) determines the effects of the drug. We detected BRAF splicing variants in tumors collected at the time of progression on PLX4032, but not in matched pre-treatment tumors, supporting the clinical relevance of this mechanism and its potential utility as a therapeutic target. These data support the model that inhibition of ERK signaling by RAF inhibitors is dependent on levels of RAS-GTP too low to support RAF dimerization and identify a novel mechanism of acquired resistance in patients: expression of splicing isoforms of BRAF(V600E) that dimerize in a RAS-independent manner. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr PR-1.

Over 50% of melanomas harbor activating mutations in BRAF, most commonly BRAF(V600E). Profound clinical activity has been observed with RAF inhibitors, including vemurafenib, which is now FDA approved for the treatment of patients with advanced melanomas whose tumors harbor a BRAF(V600E) mutation. RAF inhibitors affect ERK signaling in a mutation-specific manner: they inhibit ERK signaling in cells with BRAF(V600E), but paradoxically activate ERK signaling in cells with wild-type BRAF. We recently elucidated the mechanism of this phenomenon: activation of RAS promotes the dimerization of members of the RAF family. At non-saturating concentrations, binding of an ATP-competitive RAF inhibitor to one member of the dimer inhibits it, while also causing its transition to the activated state. This is associated with the allosteric transactivation of the other, non-drug bound member of the dimer. This leads to an overall increase in RAF specific activity and induction of ERK signaling. In BRAF(V600E) melanomas, RAS-GTP levels are low, BRAF(V600E) is found primarily as a monomer and RAF inhibitors effectively inhibit active BRAF(V600E) monomers. This model of RAF transactivation by RAF inhibitors predicts that any molecular lesion that enhances RAF dimerization will promote resistance to RAF inhibitors. To investigate mechanisms of acquired resistance to RAF inhibitors, we treated a sensitive BRAF(V600E) expressing melanoma cell line (SKME239) with the RAF inhibitor vemurafenib for 8 weeks and selected for resistant clones. We found that a subset of cells resistant to vemurafenib expressed a splicing variant form of BRAF(V600E) that lacked exons 4-8, a region that encompasses the RAS-binding domain. This form of BRAF(V600E) had a size of approximately 61KD (p61BRAF(V600E)). p61BRAF(V600E) exhibited enhanced dimerization as compared to full length BRAF(V600E) in cells with low levels of RAS-GTP. Ectopic expression of p61BRAF(V600E) conferred resistance to the RAF inhibitor. Moreover, a mutation that disrupts dimerization of p61BRAF(V600E) restored its sensitivity to vemurafenib. In tumors from patients that relapsed on vemurafenib we identified various splicing variants of BRAF(V600E), all of them lacking the RAS-binding domain (6/19). Disease progression samples obtained from a mutually exclusive subset (4/19) of the same group of patients harbored activating mutation in NRAS. We report the first RAF-inhibitor resistance mechanism that involves a structural change in BRAF and the first kinase inhibitor resistance mechanism that involves expression of aberrant splicing variants of the drug target. Tumors resistant to RAF inhibitors resulting from increased RAF dimerization retain sensitivity to inhibitors of downstream effectors of RAF such as MEK. Therefore, MEK inhibitors, if used in combination with RAF inhibitors, may delay or prevent the onset of this mechanism of resistance. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr LB-428. doi:1538-7445.AM2012-LB-428

Tumors with mutant BRAF(V600E) are dependent on ERK signaling for their growth. Unlike MEK inhibitors that suppress ERK signaling in all cells, RAF inhibitors suppress ERK signaling in cells with mutant BRAF, but paradoxically enhance ERK signaling in tumor and normal cells with wild-type BRAF. The mechanistic basis of this phenomenon lies in the interaction of these drugs with RAF dimers. Active RAS promotes homo- and heterodimerization of RAF. Binding of drug to one protomer in the RAF dimer transactivates the other. In BRAF(V600E) tumors, RAS is not activated, RAF is a monomer and thus transactivation is minimal. In these tumors, RAF inhibitors bind to and inhibit the RAF monomer and effectively suppress ERK signaling. Furthermore, because RAF inhibitors do not inhibit ERK signaling in other cells, the model predicts that they would have a higher therapeutic index and greater antitumor activity than MEK inhibitors. Indeed, RAF inhibitors have shown remarkable clinical activity in patients with melanomas that harbor mutant BRAF(V600E), however, resistance invariably develops. The mechanism of action of these drugs suggests that drug resistance could result from any lesion that increases RAF dimerization. These could include mechanisms that increase RAS activation in the tumor cells or lesions that cause RAF to dimerize in a RAF-independent manner. Indeed, mutations in NRAS and activation of receptor tyrosine kinases have now been associated with innate or acquired resistance to RAF inhibitors. We have found that aberrantly spliced forms of BRAF(V600E) are associated with resistance to RAF inhibitors in preclinical models and in patients that relapsed on the RAF inhibitor vemurafenib (PLX4032). These variants lack the RAS-binding domain, constitutively homodimerize in cells with low levels of RAS activity and confer resistance to the drug. Thus, RAF dimerization emerges as a common theme in RAF-inhibitor resistance mechanisms. Since a large part of resistance to RAF inhibitors is attributable to attenuation of the ability of the drug to inhibit RAF, one would predict that such tumors would retain at least partial sensitivity to inhibitors of downstream effectors of RAF such as MEK or ERK. Therefore, combinatorial approaches targeting upstream or downstream of RAF in addition to RAF inhibitors may delay or prevent the onset of these mechanisms of resistance.

MicroScale: miniaturizing cell arrays for macrogenetic screens

Background: RAF and EGFR inhibitor combinations have shown promising clinical activity in patients with BRAF V600E colorectal cancer, but resistance invariably develops. Methods: To define mechanisms of resistance to RAF/EGFR inhibition, we analyzed pre-treatment and disease progression samples from patients with BRAF mutant colorectal cancer using our custom next-generation sequencing platform, MSK-IMPACT (Integrated Mutation Profiling of Actionable Cancer Targets). Cell line and patient derived xenograft models were generated to investigate the effects of the genetic alterations identified at disease progression. Results: We identified alterations predicted to result in RAF dimerization and re-activation of ERK signaling in seven of eight post-progression tumors. These alterations included an intragenic deletion in BRAF encompassing the RAS binding domain, KRAS and NRAS mutations, and KRAS, NRAS, and BRAF amplifications. Notably, these alterations were rapidly enriched in the presence of drug treatment but selected against in the absence of drug exposure. A novel RAF inhibitor, that equipotently inhibits RAF mutant monomers and dimers, effectively suppressed ERK signaling in the RAF/EGFR resistant colorectal tumors at drug doses that did not inhibit ERK signaling in RAS/RAF wild-type cells. Treatment with this novel RAF inhibitor was able to overcome resistance in cell line and patient derived xenograft models. Conclusions: Our data suggest that alterations that promote RAF dimers, which are resistant to first generation RAF inhibitors, are a unifying mechanism of RAF/EGFR inhibitor resistance in patients with BRAF mutant colorectal cancer. These results support the testing of RAF dimer inhibitors in combination with EGFR inhibitors in patients with BRAF V600E colorectal cancer. Citation Format: Rona Yaeger, Zhan Yao, David M. Hyman, Jaclyn F. Hechtman, Efsevia Vakiani, Andrea Cercek, Jose Baselga, Elisa DeStanchina, Leonard Saltz, Michael F. Berger, David B. Solit, Neal Rosen. RAF kinase dimerization mediates clinical acquired resistance to RAF/EGFR inhibition in BRAF V600E colorectal cancer. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr C174.

Drugs that compete with ATP for binding to RAF selectively inhibit ERK signaling in tumor cells with BRAF(V600E) mutation. In BRAF wild-type cells, these compounds induce ERK signaling. RAF family members form dimers in a RAS-dependent manner. In BRAF wild-type cells, binding of RAF inhibitors to one member of the dimer transactivates the other, non-bound member. In contrast, in tumors with BRAF(V600E) mutation, RAS activity is too low to support dimer formation and the drug thus inhibits ERK signaling. This property of RAF inhibitors such as PLX4032 is believed to account for the wide therapeutic index of this drug and its remarkable clinical activity in patients with BRAF(V600E) melanomas. However, despite the initial effectiveness of PLX4032, resistance invariably develops. The dimerization model predicts that induction of RAS activity in the tumor will confer resistance. RAS mutation and activation of receptor tyrosine kinases (RTK) including PDGFRß and IGF1R have been implicated, as has induction of expression of the COT kinase, which drives RAF-independent ERK signaling. In order to identify novel mechanisms of RAF-inhibitor resistance, we generated PLX4032-resistant cell lines by prolonged exposure of sensitive BRAF mutant melanoma cells to the RAF inhibitor PLX4032. Analysis revealed that resistance of these clones was associated with failure of the drug to inhibit ERK signaling. PLX4032-resistant cells remained dependent on ERK signaling, as growth was inhibited by a MEK inhibitor, albeit at slightly higher doses. We did not detect RAS mutations, upregulation of receptor tyrosine kinases (RTK), or of other MEK kinases (CRAF, COT, Mos) in these cells. In a subset of resistant clones, we identified the expression of a variant form of BRAF(V600E), with an in-frame deletion within the N-terminal regulatory domain and a molecular weight of about 61KD (p61BRAF(V600E)). The N-terminal domain of RAF has been shown to negatively regulate the C-terminal catalytic domain, at least in part, by preventing dimerization. When expressed in 293H cells with intrincically low levels of RAS.GTP, dimerization of p61BRAF(V600E) is elevated compared to that of full-length BRAF(V600E). Expression of p61BRAF(V600E) in 293H cells rendered ERK signaling insensitive to inhibition by PLX4032, whereas a mutation in the dimerization domain of p61BRAF(V600E) restored sensitivity. Thus, in addition to BRAF gatekeeper mutations, there are at least three potential mechanisms that confer resistance by blunting the inhibition of RAF kinase by the inhibitor, all of which promote RAF dimerization: increased RAS.GTP due to RAS mutation or upstream (RTK) activation, RAF overexpression and deletion of the RAS binding domain within the N-terminus of RAF. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr LB-419. doi:10.1158/1538-7445.AM2011-LB-419

Distinct Binding Preferences between Ras and Raf Family Members and the Impact on Oncogenic Ras Signaling.

Constitutive activation of the RAS, RAF, mitogen-activated protein kinase (MEK), and extracellular signal regulated kinase (ERK) pathway is a common finding in many human cancers. Activating mutations in K-RAS or B-RAF constitute over 30% of all mutations in human tumors. It has been demonstrated that RAF inhibitors selectively inhibit B-RAFV600E tumors and not RAS mutant tumors, despite functioning as one of the key effector enzymes downstream of RAS and upstream of MEK. Recent Phase I clinical data with RAF inhibitors has demonstrated a significant response rate (81% response rate) in metastatic melanoma patients with B-RAFV600E-positive tumors (N Engl J Med. 2010 363:809-19). In contrast, the response rates in B-RAFV600E-positive colon tumors have been relatively low. In this work we examine the molecular mechanism of EGFR pathway activation as a mechanism of resistance by comparing the effects of RAF inhibitors against a panel of B-RAFV600E, mutant K-RAS, and WT RAS/RAF tumor lines in the presence and absence of EGF stimulation. We demonstrate that in B-RAF WT lines, pathway induction is observed for RAF inhibitors as evidenced by increases in phospho-MEK and phospho-ERK levels. RAF-inhibitor induced pathway activation in this setting mimics the kinetics and mechanism of EGF-induced pathway stimulation where RAS-GTP levels, C-RAF kinase activity and pERK levels are elevated. Upon stimulation with EGF or serum in B-RAF WT or K-RAS mutant lines, RAF inhibitor induced pathway activation is attenuated. In contrast in B-RAFV600E tumor lines, RAF inhibitors effectively block the MAPK pathway under basal conditions but become ineffective in a subset of B-RAFV600E lines when cells are stimulated with EGF. Additional studies across a broader panel of colon and melanoma mutant B-RAF lines demonstrate that B-RAFV600E lines with high basal levels of EGFR tend to be more resistant to selective B-RAFV600E inhibitors. Immunohistochemistry of colorectal tumor patient samples with either wildtype B-RAF or B-RAFV600E illustrates that colorectal B-RAFV600E tumors are associated with a higher percentage of membrane EGFR. Taken together, these results provide insight into the therapeutic utility of MAPK pathway inhibitors, and emphasize not only the importance of targeting defined genetic backgrounds in cancer treatment but the value in assessing other key pathway markers (such as EGFR) in B-RAFV600E-positive tumor subsets as a factor in sensitivity or resistance to therapy. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 258. doi:10.1158/1538-7445.AM2011-258

Despite recent successes in drug development, metastatic melanoma remains a challenging disease. Efforts to design new targeted therapies as well as improve efficacy with combinations of existing inhibitors have yielded increases in progression-free and overall survival. However, as with many targeted therapies across the cancer spectrum, habitual treatment often leads to drug resistance and progression necessitating alternative treatment regimens. More recently, excitement surrounding the success of immunotherapies including ipilimumab, nivolumab, and pembrolizumab, which facilitate long-term responses via immune-mediated cytotoxicity, has been tempered by toxicity limitations and comparatively low response rates. An efficacious and durable treatment for patients suffering from metastatic melanoma is an unmet need. While improvements with immunotherapy regimes are being intensely investigated, studies with targeted inhibitors remain highly important due to their relatively high response rates and speed to debulk systemic tumor burden. One such study that investigates the mode of action of various RAF inhibitors has recently been published in Cancer Cell by Karoulia et al. Karoulia et al. present a unified model of the biochemical nuances of a number of clinical and preclinical RAF inhibitors in various mutational contexts. Prominently featured in this work is the conformational shift of the BRAF αC helix induced by inhibitor loading. The position of the αC helix in BRAF has long been understood as a critical determinant of kinase function and dimerization. Dimerization of RAF family members is known to propagate growth signals downstream of active RAS signaling; however, mutations in BRAF such as the V600E substitution facilitate monomeric BRAF kinase activity. Karoulia et al. utilize existing and newly generated BRAF crystal structure modeling to categorize RAF inhibitors as either αC helix IN-inducing or OUT-inducing inhibitors which shift the structural position of the αC helix either in toward the kinase or out into space, respectively (Figure 1). The group demonstrates that as designed, mutant BRAF targeting αC helix OUT-inducing inhibitors, including vemurafenib, dabrafenib, and PLX7904, fail to suppress ERK signaling in cells expressing wild-type BRAF. PLX7904 (‘Paradox Breaker’—PB) is a preclinical RAF inhibitor designed to specifically target mutant V600 BRAF while limiting paradoxical activation of ERK1/2 signaling in wild-type BRAF-containing tissues (Zhang et al., 2015). In addition, Karoulia et al. show that vemurafenib and dabrafenib are ineffective against mutant BRAF-harboring cells that possess NRAS binding deficient mutant BRAF splice variants and mutant BRAF thyroid and colorectal cell lines with high RAS activity. In contrast, many of the αC helix IN inhibitors, such as AZ-628, TAK-632, LY3009120, and GDC-0879, do not show selectivity to BRAF V600E and are able to suppress ERK signaling and growth in these settings, as well as decrease the proclivity for resistance. Resistance to FDA approved αC helix OUT inhibitors often involves mechanisms that favor BRAF dimerization (Yao et al., 2015). Based on crystal structure modeling of BRAF/BRAF homodimers, the group postulates that the difference in response and penchant of resistance between IN and OUT inhibitors reflects the OUT displacement of the αC helix imparting allosteric hindrance onto an adjacent protomer, rendering it unable to load inhibitor. Utilizing both a chemically induced system and an immunoprecipitation assay, Karoulia et al. demonstrate that modest levels of the αC OUT inhibitor vemurafenib do not alter BRAF/CRAF dimerization in mutant NRAS SKMEL2 and mutant KRAS HCT116 cells. Increased dosing seems to facilitate heterodimerization. It is of note that the magnitude of high-dose vemurafenib-induced dimerization is far lower than that observed after treatment with αC IN inhibitors. The differences in dimerization observed with αC IN and αC OUT inhibitors may align with work from other groups’ crystal structure analysis suggesting that αC OUT inhibitor loading induces a conformational change of the bound protomer that reduces homophilic binding affinity for a second protomer (Thevakumaran et al., 2015). In contrast to the effects of αC IN inhibitors and to a lesser extent than the OUT inhibitor vemurafenib, BRAF/CRAF heterodimerization in the presence of PB is reduced at modest levels and is further repressed with higher dosing. On closer examination of the conformational changes induced by RAF inhibitor loading, the group observed changes of BRAF residue R506 at the base of the αC helix. Karoulia et al. found that the αC IN-inducing inhibitors stabilized R506 in the ‘IN’ position. The αC OUT inhibitors vemurafenib and dabrafenib displayed a shift of R506 toward the ‘OUT’ direction, while PB induced the most extreme outward movement in R506 (Figure 1). R506 was previously found to be a critical residue in BRAF/CRAF heterodimerization. This finding supports the notion that binding of RAF inhibitor induced conformational changes in the αC helix and R506 orientation may regulate binding of RAF to RAS-GTP and consequently RAF dimerization. Karoulia et al. show that PB is less potent in suppressing ERK signaling in cells containing splice variants of mutant BRAF, compared to cells expressing monomeric mutant BRAF. These results contrast previous reports showing PB to be a more potent inhibitor than vemurafenib in cell lines expressing other splice variants of BRAF V600E (Basile et al., 2014) and perhaps in astrocytoma cells expressing a constitutively dimerizing BRAF fusion protein. PB may suppress dimeric BRAF V600E activity in cells better than vemurafenib when used at similar concentrations, as it is a much more potent BRAF V600E inhibitor in vitro (Zhang et al., 2015) and it does not promote RAF/RAS-GTP interaction and RAF dimerization (Karoulia et al., 2016). It should be noted that Karoulia et al. only measured BRAF/CRAF heterodimerization. Heterodimerization may be important in wild-type BRAF malignancies and in paradoxical pathway activation but less so in the context of mutant BRAF driven tumors. Furthermore, CRAF is dispensable in vemurafenib-resistant mutant BRAF cells harboring splice variants that require dimerization (Poulikakos et al., 2011). Perhaps a key determinant of the response to OUT inhibitors is the extent of BRAF/BRAF homodimerization achieved during treatment; direct measurement of the levels of BRAF/BRAF homodimerization may shed light on the unique mechanism of PB action. This work highlights the importance of disease genotype and peripheral pathway activation when deciding to administer αC IN versus OUT RAF inhibitors. Data presented by Karoulia et al. demonstrate strong sensitivity in all cell types tested with the pan-RAF αC IN inhibitors. This is most likely due to this class of inhibitors blocking both BRAF and CRAF kinase activity regardless of upstream RAS activation or RAF level mutation. Consequently, αC IN inhibitors should be considered for wild-type BRAF/wild-type NRAS melanoma, mutant NRAS melanoma, and mutant BRAF malignancies associated with high RAS activity such as colorectal or thyroid cancers. On the other hand, αC OUT inhibitors should be considered for BRAF V600 mutants, which signal as monomers as described by Yao et al. (2015). Toxicity due to paradoxical activation elicited by the αC OUT inhibitors is of concern; however, PB is a promising new RAF inhibitor in early clinical trials designed to reduce these side effects while still affording high therapeutic index due to mutant BRAF selectivity. ‘Karoulia et al. demonstrate the power of pan-RAF inhibitors’ The authors of this manuscript are supported by the National Cancer Center (E.J. Hartsough) and the National Institute of Health, National Cancer Institute (F30CA203314, M.J. Vido).

Tumours with mutant BRAF are dependent on the RAF-MEK-ERK signalling pathway for their growth. We found that ATP-competitive RAF inhibitors inhibit ERK signalling in cells with mutant BRAF, but unexpectedly enhance signalling in cells with wild-type BRAF. Here we demonstrate the mechanistic basis for these findings. We used chemical genetic methods to show that drug-mediated transactivation of RAF dimers is responsible for paradoxical activation of the enzyme by inhibitors. Induction of ERK signalling requires direct binding of the drug to the ATP-binding site of one kinase of the dimer and is dependent on RAS activity. Drug binding to one member of RAF homodimers (CRAF-CRAF) or heterodimers (CRAF-BRAF) inhibits one protomer, but results in transactivation of the drug-free protomer. In BRAF(V600E) tumours, RAS is not activated, thus transactivation is minimal and ERK signalling is inhibited in cells exposed to RAF inhibitors. These results indicate that RAF inhibitors will be effective in tumours in which BRAF is mutated. Furthermore, because RAF inhibitors do not inhibit ERK signalling in other cells, the model predicts that they would have a higher therapeutic index and greater antitumour activity than mitogen-activated protein kinase (MEK) inhibitors, but could also cause toxicity due to MEK/ERK activation. These predictions have been borne out in a recent clinical trial of the RAF inhibitor PLX4032 (refs 4, 5). The model indicates that promotion of RAF dimerization by elevation of wild-type RAF expression or RAS activity could lead to drug resistance in mutant BRAF tumours. In agreement with this prediction, RAF inhibitors do not inhibit ERK signalling in cells that coexpress BRAF(V600E) and mutant RAS.

MAPK activation through KRas, NRas or BRaf mutation occurs in approximately 70% of colorectal cancer patients. Due to their epithelial origin, colorectal tumors generally have high levels of EGFR expression and activation. EGFR therapies such as cetuximab are effective for treatment of a subset of colorectal cancer, particularly patients with wild type (WT) KRas. EGFR signaling is also recently identified as a key resistance mechanism in BRaf mutant colorectal cancer to BRaf inhibitors. In this study, we have genetically characterized 78 patient-derived xenograft (PDX) models of colorectal tumors, and conducted an “n = 1” (single mouse per treatment group) trial in these PDX models with cetuximab, LSN3074753, a pan-Raf and Raf dimer inhibitor, and their combination in collaboration with Oncotest GmbH and Champions Oncology. Among these 78 PDX models, 42 (53.8%) have a KRas mutation, 12 (15.4%) have BRaf V600E or an atypical BRaf mutation, and 26 (33.3%) are WT KRas and BRaf. Consistent with clinical results, cetuximab is primarily active in WT KRas and BRaf PDX models, with disease control rate (DCR) of 53.8% (14/26) in this subgroup. These results suggest that the mouse n = 1 PDX trial paradigm could reliably predict clinical results. For pan-Raf and Raf dimer inhibitor LSN3074753, it is active in a subset of PDX models, particularly those with BRaf or KRas mutation(s), with DCR of 21.2% among models with a KRas or BRaf mutation. Importantly, a synergistic effect is observed when cetuximab and LSN3074753 are combined for treatment of these 78 PDX models. The overall DCR in the combination arm is 50% (39/78), while cetuximab or LSN3074753 alone has an overall DCR of 24 or 18%, respectively. Further statistical analyses reveal that BRaf mutations including V600E or other atypical mutations (G469E, G76E, G596V, G203V, etc) are the best predictor of combination synergy, and are significantly associated with synergistic effect with a p value of 0.004. In models with BRaf mutations, the combination arm has a DCR of 50% (6/12), whereas cetuximab or LSN3074753 alone has a DCR of 8.3 or 17%, respectively. BRaf or KRas mutations are also significantly associated with combination synergy with p value of 0.01. Among 42 KRas mutation models, LSN3074753 or cetuximab alone has a DCR of 21.4 or 16.7%, and the combination arm has a DCR of 43%. Overall, these results indicate that combination of EGFR and Raf inhibition by cetuximab and a pan-Raf inhibitor has the potential for treatment of colorectal cancer patients with BRaf or KRas mutation. Citation Format: Yung-mae M. Yao, Gregory P. Donoho, Philip W. Iversen, Yue Wang Webster, Yong Gang Yue, James R. Henry, Gregory D. Plowman, Sheng-Bin Peng. Mouse PDX Trial Suggests Combination Efficacy of Raf and EGFR Inhibition in Colorectal Cancer with BRaf or KRas mutation. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr LB-004. doi:10.1158/1538-7445.AM2015-LB-004

Recent years have seen dramatic clinical advances in targeting the ERK pathway with the FDA approval of several selective inhibitors of RAF and MEK1–5. Clinical trials of these agents indicate that near complete inhibition of pathway signaling is necessary to effectively inhibit tumor growth6. Targeted approaches have been most successful in BRAF V600E tumors because of the relatively wide therapeutic index of RAF inhibitors (vemurafenib and dabrafenib), which inhibit signaling in cells with BRAF V600 mutants but cause “paradoxical activation” of ERK signaling in normal cells that are wild-type for RAF kinases. In contrast, MEK inhibitors (trametinib and cobimetinib) suppress ERK signaling in all cells, and their clinical activity has been limited by a narrow therapeutic index. Combining these agents – with RAF and MEK inhibitors for BRAF V600E melanoma or lung cancer and with RAF, MEK, and EGFR inhibitors for BRAF V600E colorectal cancer – to profoundly inhibit ERK signaling has led to improved antitumor activity7. Combination therapy has also been shown to offset the toxicities caused by RAF inhibitors, such as the development of keratoacanthomas and squamous cell carcinoma, resulting from paradoxical ERK activation with these agents. The role of RAF inhibitors to offset MEK inhibitor toxicity and the need for dose intensity to modulate opposing toxicities is less clear. Recent observations in clinical trials have suggested that RAF inhibitors offset dermatologic toxicity from MEK or EGFR inhibitors. In the phase III trial of trametinib in melanoma, grade 3 or 4 acneiform dermatitis occurred in 8% of trametinib-treated patients, whereas in the phase III trial of the combination of dabrafenib and trametinib, no patient had grade 3 or 4 acneiform dermatitis8,9. Combinations of RAF and EGFR inhibitors have also had a lower incidence of acneiform rash than seen with EGFR inhibitors alone7,10,11. This is also likely due to the opposite effects of RAF and EGFR inhibitors on MEK activation in normal cells. However, the doses of RAF inhibitors needed for these clinically opposing effects and how these doses compare to clinically efficacious doses have not been studied. We now report the course of a patient with BRAF V600E CRC treated with dabrafenib, trametinib, and panitumumab in a phase II clinical trial, and within this patient characterize the effect on toxicities of different dose levels of these agents. Further, we find that, within the clinical dose range, there is a RAF inhibitor dose that is an inflection point for the toxicity and efficacy of this regimen.

BRAF V600E drives tumors by hyperactivating ERK signaling, is activated in a RAS-independent manner, and signals as a monomer. Selective inhibitors of RAF potently inhibit BRAF V600E and downstream ERK signaling, and have been associated with impressive clinical responses in patients whose tumors harbor V600E mutations in BRAF. Almost all patients, however, ultimately develop resistance. To date, acquired resistance mechanisms validated in patient samples have included splice forms of BRAF, mutations in NRAS, and BRAF amplification, all of which promote formation of RAF dimers, and mutations in MEK. Second-site point mutations in BRAF that promote dimer formation and are associated with acquired resistance have not previously been identified in RAF inhibitor-resistant tumors. Here, we report a complete response followed by clinical progression of a BRAF V600E-mutant pediatric brain tumor treated with the RAF inhibitor dabrafenib. We identified a second-site mutation in BRAF at the time of progression that was not present in the pretreatment tumor. Further study indicated the acquired mutation is in cis with V600E. We demonstrate that the acquired resistance mutation induces RAF dimer formation, and is sufficient to confer resistance to dabrafenib. Moreover, we showed that ERK signaling activated by the double mutation is sensitive to MEK inhibition or to the RAF plus and MEK inhibitor combination. We also find that two novel RAF dimer inhibitors may overcome resistance mediated by the novel mutation in BRAF. This is the first study, to our knowledge, that demonstrates the emergence of a confirmed second-site mutation in BRAF V600E and validates that this mutation is responsible for acquired resistance in an initially responding patient treated with dabrafenib. The mechanism of acquired resistance can be abrogated by the novel RAF dimer inhibitors. Our data confirm that a novel class of RAF dimer inhibitors are active against an acquired resistance mutation, suggesting an improved personalized treatment option for patients who harbor similar second-site mutations in BRAF. Citation Format: Jiawan Wang, Zhan Yao, Philip Jonsson, Amy Allen, Alice Can Ran Qin, David Pisapia, Neal Rosen, Barry S. Taylor, Christine A. Pratilas. A second-site mutation in BRAF confers resistance to RAF inhibition in a BRAF V600E-mutant brain tumor [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2017 Oct 26-30; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2018;17(1 Suppl):Abstract nr A129.

Highlights from Recent Cancer Literature

NG04: Clinical acquired resistance to RAF inhibitor combinations in BRAF mutant colorectal cancer through MAPK pathway alterations