The CDK4/6 Inhibitor LY2835219 Overcomes Vemurafenib Resistance Resulting from MAPK Reactivation and Cyclin D1 Upregulation

Published OnlineFirst August 13, 2014; DOI: 10.1158/1535-7163.MCT-14-0257

Molecular Cancer Small Molecule Therapeutics Therapeutics

Vipin Yadav1, Teresa F. Burke1, Lysiane Huber1, Robert D. Van Horn1, Youyan Zhang1, Sean G. Buchanan1, Edward M. Chan2, James J. Starling1, Richard P. Beckmann1, and Sheng-Bin Peng1

Abstract B-RAF selective inhibitors, including vemurafenib, were recently developed as effective therapies for melanoma patients with B-RAF V600E mutation. However, most patients treated with vemurafenib eventually develop resistance largely due to reactivation of MAPK signaling. Inhibitors of MAPK signaling, including MEK1/2 inhibitor trametinib, failed to show signiﬁcant clinical beneﬁt in patients with acquired resistance to vemurafenib. Here, we describe that cell lines with acquired resistance to vemurafenib show reactivation of MAPK signaling and upregulation of cyclin D1 and are sensitive to inhibition of LY2835219, a selective inhibitor of cyclin-dependent kinase (CDK) 4/6. LY2835219 was demonstrated to inhibit growth of melanoma A375 tumor xenografts and delay tumor recurrence in combination with vemurafenib. Furthermore, we developed an in vivo vemurafenib-resistant model by continuous administration of vemurafenib in A375 xenografts. Consistently, we found that MAPK is reactivated and cyclin D1 is elevated in vemurafenib-resistant tumors, as well as in the resistant cell lines derived from these tumors. Importantly, LY2835219 exhibited tumor growth regression in a vemurafenib-resistant model. Mechanistic analysis revealed that LY2835219 induced apoptotic cell death in a concentration-dependent manner in vemurafenib-resistant cells whereas it primarily

mediated cell-cycle G1 arrest in the parental cells. Similarly, RNAi-mediated knockdown of cyclin D1 induced signiﬁcantly higher rate of apoptosis in the resistant cells than in parental cells, suggesting that elevated cyclin D1 activity is important for the survival of vemurafenib-resistant cells. Altogether, we propose that targeting cyclin D1–CDK4/6 signaling by LY2835219 is an effective strategy to overcome MAPK-mediated resistance to B-RAF inhibitors in B-RAF V600E melanoma. Mol Cancer Ther; 13(10); 2253–63. 2014 AACR.

Introduction developed within 5 to 7 months (4, 6). Thus, the emer- B-RAF is the most commonly mutated driver onco- gence of resistance remains a considerable therapeutic gene in melanoma, with activating mutations in codon challenge to achieve durable responses and prolonged 600 occurring in almost 50% of the patients (1, 2). survival in these patients. Treatment with B-RAF selective inhibitors, such as A variety of molecular mechanisms are identified to be vemurafenib or dabrafenib, has demonstrated signifi- involved in resistance to B-RAF inhibition. The most cant benefit in melanoma patients with B-RAF V600E common resistant mechanism is MAPK pathway reacti- mutation, with extended patient progression-free vation, which is caused by genetic mutation of MEK or survival and median overall survival compared with different Ras isoforms (7–9), upstream activation of recep- chemotherapy (3–6). However, these responses were tor tyrosine kinases (RTK) such as FGFR3 and c-Met (10, relatively short-lived, and drug resistance generally 11), expression of B-RAF V600E splice variants that dimer- ize in presence of the B-RAF inhibitor (12), amplification of B-RAF (13, 14), and upregulation of MAP3Ks such as COT 1Oncology Discovery Research, Lilly Research Laboratories, Indianapolis, or C-RAF (15, 16). Alternatively, activation of MAPK- Indiana. 2Oncology Business Unit, Eli Lilly and Company, Indianapolis, redundant pathways such as PI3K/Akt as a consequence Indiana. of PTEN loss (17) or overexpression of RTKs such as Note: Supplementary data for this article are available at Molecular Cancer PDGFRb and IGF1R have also been reported to induce Therapeutics Online (http://mct.aacrjournals.org/). resistance in B-RAF V600E melanoma (7, 18, 19). In addi- Corresponding Authors: Richard P. Beckmann, Oncology Discovery tion, secretion of growth factors such as HGF or FGF has Research, Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, IN 46285. Phone: 317-276-4285; Fax: 317-651-6346; E-mail: also been implicated in resistance to B-RAF inhibition (10, [email protected]; and Sheng-Bin Peng. Phone: 317-433- 11, 20, 21). Although the resistant mechanisms are fre- 4549; Fax: 317-276-1414; E-mail: [email protected] quently associated with MAPK reactivation, treatment doi: 10.1158/1535-7163.MCT-14-0257 with the MEK inhibitor trametinib, or with trametinib 2014 American Association for Cancer Research. plus dabrafenib, has not been very effective in patients

www.aacrjournals.org 2253

Yadav et al.

who have previously failed B-RAF inhibitor (22), suggest- performs cell line characterizations. All these cells were ing that subsequent targeting of MAPK signaling alone is passaged for fewer than 2 months after which time new not sufficient. Therefore, despite recent advances in the cultures were initiated from vials of frozen cells. Charac- clinic, drug resistance upon selective B-RAF inhibition terization of the cell lines was done by a third-party remains a considerable therapeutic challenge in clinic. vendor (RADIL, which included profiling by PCR) for Constitutive activation of cyclin-dependent kinases contamination by various microorganisms of bacterial (CDK) and deregulation of cell cycle are common features and viral origin. As a result, no contamination was across several cancer types, including melanoma. detected. The samples were also verified to be of human INK4a P16 , a tumor suppressor gene and a negative regu- origin without mammalian interspecies contamination. lator of CDK4, is deleted in 38% of melanoma (2, 23). In The alleles for 9 different genetic markers were used to addition, germline mutations and amplification of CDK4 determine that the banked cells matched the genetic gene have been identified in melanoma, which leads to profile that has been previously reported. unrestricted CDK4 activity and increased cell prolifera- All cells were maintained in DMEM (Thermo Scientific) tion (2, 24). In general, regulation of cell-cycle entry in supplemented with 10% FBS (Invitrogen). B-RAF selective proliferating adult mammalian cells is controlled by D- inhibitor vemurafenib and CDK4/6 dual inhibitor cyclins which bind and activate CDK4 and CDK6 to LY2835219 were synthesized by Eli Lilly and Company. promote phosphorylation of retinoblastoma (Rb) protein The mesylate salt of LY2835219 (LY2835219.CH4O3S) was in vitro and G1 to S transition (25). The RAS–MAPK pathway is used in all the studies. All siRNAs were obtained known to control cell-cycle entry via upregulation of from Dharmacon (OnTargetPlus SiRNA). siRNA transfec- cyclin D1 in several cell types (26, 27). Inhibition of MAPK tions were performed using Lipofectamine RNAiMAX signaling by B-RAF inhibitors decreases cyclin D1 expres- transfection reagent (Invitrogen) as per manufacturer’s sion and upregulates CDK inhibitor p27KIP1 levels, thus instructions. Cyclin D1 and nonspecific control shRNA blocking cell-cycle entry in B-RAF V600E melanoma lentiviral particles were obtained from Sigma-Aldrich (26, 27). Overexpression of cyclin D1 is linked to resistance (Mission shRNA). Cells were transduced with the virus to B-RAF inhibition (28). CCND1 is amplified in 11% of for 72 hours as per manufacturer’s instructions. Antibo- melanoma, including 17% of B-RAF V600E melanoma, dies against phospho-ERK1/2 (4370), phospho-Akt1 thus suggesting a potential role of cyclin D1 in intrinsic (S473) (4060), phospho-MEK (9154), cyclin D1 (2978), resistance to B-RAF inhibitors (2, 28). However, the role of cyclin D2 (3741), phospho-Rb (S807/811) (8516), p27 cyclin D1 in acquired resistance to vemurafenib has not (2552), cleaved PARP (9541), and cleaved caspase-3 been described, and the therapeutic value of targeting (9664) were purchased from Cell Signaling Technology. cyclin D1/Rb axis to overcome vemurafenib resistance Antibodies against B-RAF (sc-166, Santa Cruz), GAPDH has not been explored. (Santa Cruz), tubulin (ab7291, AbCam), phospho-histone In this study, we have generated multiple in vitro cell H3 (S10) (Millipore), and phospho-Rb (S780) (BD Bio- lines and an in vivo model of resistance to vemurafenib and sciences) were obtained from the indicated companies. discovered that MAPK reactivation and cyclin D1 elevation are common in these resistant models. We describe Generation of vemurafenib-resistant cell lines that cyclin D1 is an important mediator of vemurafenib A375R1, A375R3, M14R, and SH4R models with resistance and provides a potential therapeutic target to acquired resistance to vemurafenib were generated by overcome resistance to B-RAF inhibition in B-RAF V600E treating respective parental cells with gradually increasing melanoma. Using these in vitro and in vivo models, we concentrations of vemurafenib, up to 2 mmol/L, as previ- show that cyclin D1 is generally elevated and functions as a ously described (10). A375R2 cells were generated by critical node for the survival of vemurafenib-resistant cells. treating A375 cells with a high concentration of vemur- We further demonstrate that LY2835219, a selective dual afenib (2 mmol/L) for 1 week, followed by culturing with CDK4/6 inhibitor currently in phase II clinic studies (29), vemurafenib (1 mmol/L) for up to 20 passages. Upon can overcome vemurafenib resistance in these resistant establishment of resistance, the inhibitor was withdrawn models. Altogether, this study sheds new light on mechan- and all of the cell lines were maintained in regular media isms of resistance to B-RAF inhibition, identifies cyclin D1 for subsequent passages. For A375R4 cells, the NRas Q61K elevation concurrent with MAPK reactivation as a com- was stably transfected into A375 cells, and single-cell clone mon resistant mechanism, and proposes targeting cyclin was selected and characterized for this study. All of the D1 through CDK4/6 inhibition by LY2835219 as an effec- models retained resistance to vemurafenib for up to 20 tive therapeutic strategy to overcome B-RAF resistance. passages (data not shown). In this study, all resistant cell lines with less than 10 passages were used for experiments. Materials and Methods Cell culture, reagents, and transfections Cell proliferation assay A375, SH4, and A2058 cells were obtained from ATCC Cells (1 103 to 3 103) were normally plated in 96- on May 7, 2012, July 2, 2012, and June 27, 2006, respec- well plates (BD Biosciences). Cells were treated the next tively. M14 cells were purchased from NCI on March 10, day for 96 hours, and then assessed for viability using 2005. Cells were stored within a central cell bank that CellTiter Glo (Promega), as per manufacturer’s

2254 Mol Cancer Ther; 13(10) October 2014 Molecular Cancer Therapeutics

Target Vemurafenib Resistance by the CDK4/6 Inhibitor LY2835219

instructions and a luminescence plate reader (Victor, dose schedules described in each study. Tumor growth Perkin Elmer). GraphPad Prism 4 software was used to and body weight were monitored over time to evaluate generate sigmoidal dose-response curves and calculate efficacy. the proliferation IC50. Generation of cell lines from xenograft tumors Caspase-3/7 activity assay Tumors were aseptically removed from the animals, Cells (5 103) were plated in 96 well plates (BD washed with cold PBS, then minced, and trypsinized in Biosciences). Cells were treated the next day for 24 to 10-cm petri dish containing 5 mL of TrypLE reagent 48 hours and then assessed for caspase-3 activity by (Invitrogen) for 15 minutes at 37C. Dislodged cells Caspase-Glo-3/7 Assay (Promega), as per manufacturer’s were collected, resuspended, and plated in DMEM instructions and a luminescence plate reader (Victor, (Thermo Scientific) supplemented with 50% FBS (Invi- Perkin Elmer). trogen) and 10% penicillin/streptomycin (Invitrogen). After 1 week, cells were switched to regular media Preparation of cell lysates and immunoblotting (DMEM þ 10% FBS). Cells and tumor tissues were lysed using either RIPA lysis buffer (Bio-Rad; cell lines) or 1% SDS solution Results (tumor lysates), containing 1 phosphatase and pro- Development of vemurafenib-resistant cell lines teinase inhibitor cocktail (Pierce). Tumor lysates were To study the resistant mechanisms of B-RAF inhibition, prepared with freshly frozen tumor samples. Approx- we generated several vemurafenib-resistant cells outlined imately 200 mg of tumor tissue was homogenized and in Table 1. A375R1, A375R3, M14R, and SH4R cells were lysed using 0.5 mL of lysis buffer. The protein concen- generated by treating their respective parental cells with tration of individual samples was determined with DC gradually increasing concentrations of vemurafenib, up to Protein Assay Kit (Bio-Rad). SDS-PAGE was performed 2 mmol/L as described previously (10). The resistant on cell lysates containing 20 mgoftotalproteinusing4% mechanism of A375R1, A375R3, and M14R cells was to 20% Novex tri-glycine gradient gels (Invitrogen). associated with RTK/RAS activation and MAPK reacti- Protein was transferred onto 0.2-mm nitrocellulose vation (10). MAPK reactivation was also observed in SH4 membranes using Trans-Blot Turbo Transfer system cells, but the underlying mechanism has not been fully (BioRad) as per manufacturer’s instructions. Proteins characterized. A375R2 cells were generated by treating were detected using the Odyssey Infrared Imaging A375 cells with a constant 2 mmol/L vemurafenib as System (Li-COR Biosciences). described (12). Consistent with the earlier reports, the resistance of A375R2 cells to vemurafenib is conferred by Flow cytometry expression of B-RAF splice variant (Supplementary Fig. Cell pellets were fixed in 70% ethanol for 30 minutes at S1A). The A375R4 cells were generated through stable 20C and then washed with PBS. Fixed cells were transfection of NRas Q61K mutant and single clone selec- stained with propidium iodide (PI)/Triton X-100 staining tion, which resulted in vemurafenib resistance (Supple- solution and incubated for 30 minutes at room tempera- mentary Fig. S1B). ture. Fixed cells were then subjected to flow cytometric analysis on the Beckman Coulter FC 500 Cytomics flow MAPK reactivation and cyclin D1 elevation in cytometer. Data were analyzed with ModFit LT 3.0 (Verity vemurafenib-resistant cells and their sensitivity to House Software). CDK4/6 inhibitor LY2835219 As shown in Table 1, all in vitro generated resistant In vivo experiments and drug administration celllines,A375R1-R4,M14R,andSH4R,demonstrated All animal studies were performed in accordance with resistance to vemurafenib. Importantly, these resistant American Association for Laboratory Animal Care insti- cells showed enhanced MAPK activation and cyclin D1 tutional guidelines. The Eli Lilly and Company Animal elevation (Fig. 1). MAPK pathway activation and loss of Care and Use Committee approved all the experimental cell-cycle control are generally the hallmarks of mela- protocols. Athymic nude female mice were inoculated noma (2). Therefore, we used a selective CDK4/6 dual with 0.2 mL of 1 107 A375 cells, prepared in a 1:1 inhibitor LY2835219 and evaluated its growth-inhibito- Matrigel to media mixture, in the hind flank region. A ry effects in a panel of melanoma cell lines that are either total of 60 mice, 8 to 10 in each group, were used for sensitive or resistant to vemurafenib mediated by compound treatments and vemurafenib-resistant model diverse mechanisms (Table 1). Interestingly, B-RAF development in each study. Vemurafenib was formulated V600E melanoma cells that are either sensitive to by dissolving in DMSO in a volume equivalent to 5% of the vemurafenib, such as A375, M14, and SH4, or resistant final formulation volume, and then the remaining volume to vemurafenib, such as A375R1-4, M14R, and SH4R, was added to the solution of 1% methylcellulose in dis- demonstrated comparable sensitivity to LY2835219 with m tilled water. CDK4/6 inhibitor LY2835219 was formulat- IC50 values ranging from 0.3 to 0.6 mol/L (Table 1). On ed in 1% HEC in 20 mmol/L phosphate buffer, pH 2.0. the contrary, B-RAF V600E mutant A2058 cells with de Treatment was administered orally (gavage) with the novo resistance to vemurafenib via MAPK-independent

www.aacrjournals.org Mol Cancer Ther; 13(10) October 2014 2255

Yadav et al.

Table 1. Overview of drug sensitivities, mutational status, and mechanisms of resistance of melanoma cell lines tested in this study

IC50, nmol/L

Vemurafenib LY2835219 Sensitivity to Cell line Mutation vemurafenib Resistance mechanism Average SD Average SD Apoptosis A375 B-RAFV600E Sensitive — 125 35 412 103 No A375-R1 B-RAFV600E Resistant FGF-mediated ERK activation 4,541 939 407 54 Yes A375-R2 B-RAFV600E Resistant p61–B-RAF splicing 3,364 1,449 361 149 Yes A375-R3 B-RAFV600E Resistant FGF-mediated ERK activation 3,035 1,781 555 132 Yes A375-R4 B-RAFV600E Resistant NRas Q61K 8,214 2,640 468 82 — M14 B-RAFV600E Sensitive — 303 97 843 224 No M14-R B-RAFV600E Resistant FGF-mediated ERK activation 5,337 1,190 538 171 Yes SH4 B-RAFV600E Sensitive — 489 131 379 45 — SH4-R B-RAFV600E Resistant Not characterized 7,775 827 374 33 — A2058 B-RAFV600E Resistant PTEN inactivation 3,586 115 1,729 237 — A375-RV1 B-RAFV600E Resistant p61–B-RAF splicing 4,413 1,808 221 67 Yes A375-RV2 B-RAFV600E Resistant Not characterized 3,619 2,553 368 188 Yes

mechanism (i.e., PTEN deletion) were relatively insen- dose schedule. Furthermore, analysis of tumor lysates sitive to LY2835219 (Table 1). showed that LY2835219 treatment significantly reduced pS780-Rb and pS10-Histone H3 levels, indicating inhibition of cell cycle and a decrease in proliferating tumor cells Antitumor effects of vemurafenib, LY2835219, and as a result of CDK4/6 inhibition (Fig. 2C). When xenograft their combination in A375 xenograft model tumors were treated with a combination of vemurafenib In vitro analysis revealed that parental B-RAF V600E and LY2835219, an additive antitumor growth effect was melanoma cells, such as A375 cells, were sensitive to both observed (Fig. 2D). vemurafenib and LY2835219. To compare their activities in vivo , we tested their antitumor effects as single agents or Development of in vivo vemurafenib-resistant model in combination in an A375 xenograft model. As demon- and efficacy of CDK4/6 inhibitor LY2835219 in this strated in Fig. 2A, vemurafenib treatment at 15 mg/kg resistant model twice daily induced significant tumor growth regression To evaluate the efficacy of LY2835219 in the vemura- in the first 2 weeks of treatment. Tumor growth was also fenib-resistant tumors, we developed an in vivo vemur- inhibited by LY2835219 in a dose-dependent fashion (Fig. afenib-resistant model as shown in Fig. 3A. Following the 2B). Statistically significant tumor growth inhibition by establishment of tumors, vemurafenib at 15 mg/kg was LY2835219 was observed at 45 or 90 mg/kg once daily administered orally twice a day. Consistent with the previous observations in this model, vemurafenib-treated mice demonstrated significant regression in tumor volume initially. We continued dosing with vemurafenib until resistance was evident. Approximately 40 to 45 days after dosing, many animals relapsed and resistant tumors A375-R2 A375-R3 A375 A375-R1 SH4-R M14 M14-R SH4 A375-R4 started to emerge. When vemurafenib-resistant tumors pMEK1/2 reached sizes approximately 600 to 1,000 mg, the animals were randomized into 2 groups, each 7 to 8 animals. pERK1/2 One group continued to be dosed with vemurafenib, and the other group was switched to CDK4/6 inhibitor pAKT LY2835219 treatment alone at 90 mg/kg once daily schedule. As demonstrated in Fig. 3A, LY2835219-treated mice Cyclin D1 demonstrated significant tumor growth regression, whereas the tumors in vemurafenib-treated mice contin- Tubulin ued to grow. Furthermore, LY2835219-mediated tumor growth inhibition was maintained upon cessation of the treatment. To rule out the possibility that the tumors had Figure 1. MAPK reactivation and cyclin D1 elevation in vemurafenib- developed a dependence on vemurafenib for continued resistant melanoma cell lines. growth as previously described (14), we repeated the

2256 Mol Cancer Ther; 13(10) October 2014 Molecular Cancer Therapeutics

Target Vemurafenib Resistance by the CDK4/6 Inhibitor LY2835219

A 1,600 5% DMSO in 1% methylcellulose, 0.2 mL, PO, BID

3 1,400 Vemurafenib,15 mg/kg, PO, BID 1,200 1,000 800 600 400 Tumor volume, mm 200 0 0 10 20 30 40 50 60 Day Rx B 3,000 1% HEC in 25 mmol/L PB pH2, 0.2 mL, PO, QD × 21 3 2,500 LY2835219, 22.5 mg/kg, PO, QD × 21 LY2835219, 45 mg/kg, PO, QD × 21 2,000 LY2835219, 90 mg/kg, PO, QD × 21 1,500

1,000 Tumor volume, mm 500

0 20151050 4540353025 Rx Day

C LY2835219

Vehicle 22.5 mg/kg 45 mg/kg 90 mg/kg

pS780-Rb

pS10-Histone H3

GAPDH

D 6,000 1% HEC in 25 mmol/L PB pH 2, 0.2 mL, PO, QD × 21 3 5,000 Vemurafenib, 10 mg/kg, PO, BID × 21 LY2835219, 45 mg/kg, PO, QD × 21 4,000 LY2835219, 45 mg/kg/ Vemurafenib, 10 mg/kg, 3,000 2,000 1,000

Tumor volume, mm 0 0 10 20 30 40 50 Rx Day

Figure 2. Antitumor activity of vemurafenib and LY2835219 in A375 xenograft model. A, antitumor activity of vemurafenib in A375 xenografts. Mice bearing subcutaneous A375 tumors were dosed with vehicle (n ¼ 8) or 15 mg/kg vemurafenib (n ¼ 10) twice daily. The y-axis is mean tumor volume SEM. B, antitumor activity of CDK4/6 inhibitor LY2835219 in A375 xenograft model. Mice bearing subcutaneous A375 tumors were dosed with vehicle (n ¼ 7), 22.5 mg/kg (n ¼ 5), 45 mg/kg (n ¼ 5), or 90 mg/kg (n ¼ 5) LY2835219 once daily for 21 days. C, dose-dependent reduction of CDK4/6 activity in A375 xenograft tumors by LY2835219. Mice bearing subcutaneous A375 tumors were dosed with vehicle, 22.5, 45, or 90 mg/kg of LY2835219 once daily. Tumors were collected 24 hours after the third dose. Tumor lysates were prepared and analyzed by immunoblotting using indicated antibodies. D, activities of LY2835219, vemurafenib, and their combination in A375 xenograft model. Mice bearing subcutaneous A375 tumors were dosed for 21 days with either vehicle (n ¼ 8), 45 mg/kg LY2835219 (n ¼ 8), 10 mg/kg vemurafenib (n ¼ 8), or combination of 45 mg/kg LY2835219 and 10 mg/kg vemurafenib (n ¼ 8). LY2835219 was dosed once daily and vemurafenib was dosed twice daily The brown line below the x-axis indicates the dosing (Rx) period in all studies. The y-axisismeantumorvolume SEM. The pairwise comparisons of each treatment versus control are all statistically significant (P < 0.001). The tests between the combination group and each single agent group are also statistically significant (P < 0.001). The P value for the 2-way interaction to determine whether combinations are different from additive is not statistically significant (P ¼ 0.657), thus indicating that the combination of these 2 agents is additive rather than synergistic.

www.aacrjournals.org Mol Cancer Ther; 13(10) October 2014 2257

Yadav et al.

A 6,000 5% DMSO in 1% methylcellulose, 0.2 mL, PO, BID × 28 Vemurafenib, 15 mg/kg, PO, BID x 119 Figure 3. Development of in vivo 5,000 Vemurafenib, 15 mg/kg, PO, BID x 92/LY2835219, 90 mg/kg, PO, QD x 28 3 vemurafenib-resistant A375 xenograft model and activity of 4,000 CDK4/6 inhibitor LY2835219 in the resistant model. A, efﬁcacy 3,000 of LY2835219 in an in vivo model of acquired resistance to 2,000 vemurafenib. Mice bearing Tumor volume, mm volume, Tumor subcutaneous A375 tumors were dosed with vehicle or 1,000 15 mg/kg vemurafenib twice daily after tumors were 0 established. Dosing was continued until the resistant Vemurafenib tumors were evident. Upon Day LY2835219 emergence of resistant tumors with size 600 to 1,000 mg, mice were randomized and dosed B Vehicle Vemurafenib-resistant tumors with 90 mg/kg LY2835219 once 1 2 3 4 5 6 7 1 2 3 4 5 6 7 8 9 10 11 a day, or continued dosing with 15 mg/kg vemurafenib twice pMEK1/2 daily. The blue or red line under the x-axis indicates the dosing pERK1/2 (Rx) period of vemurafenib or LY2835219, respectively. y Cyclin D1 The -axis is mean tumor volume SEM. B, cyclin pRb S807/ D1 elevation in vemurafenib- S811 resistant tumors. Tumor pRb S780 lysates were analyzed by immunoblotting using indicated pAkt S473 antibodies.

Tubulin

same experiment with a vehicle control arm in addition to currently being evaluated in other resistant cell lines. the continued vemurafenib arm and the LY2835219 treat- Taken together, our results indicate that MAPK reactiva- ment group. In this study, both the vehicle control group tion and upregulation of cyclin D1 are associated with and the vemurafenib group showed indistinguishable vemurafenib resistance and sensitivity to CDK4/6 and continued growth after vemurafenib withdrawal inhibition. (Supplementary Fig. S2). These results suggest that CDK4/6 inhibitor LY2835219 as a single agent is effective Cyclin D1 upregulation and vemurafenib resistance in overcoming vemurafenib resistance in this in vivo in cells generated from resistant xenograft tumors model. We further investigated the molecular mechanism To further understand the mechanism of vemurafenib behind increased sensitivity of resistant tumors to resistance and increased sensitivity to LY2835219, we LY2835219. We found hyperelevated levels of phospho- generated 2 cell lines from vemurafenib-resistant ERK, phospho-MEK, cyclin D1, and phospho-Rb (S780, tumors, hereafter referred them to as A375RV1 and S807, S811) in these vemurafenib-resistant tumors, indi- A375RV2 cells. Genotype analysis [short tandem repeat cating upregulation of MAPK and CDK activity (Fig. 3B). (STR)] confirmed that A375RV1 and A375RV2 have the This observation is consistent with previous findings that same genotypes of parental A375 cells. Compared with upregulation of cyclin D1 was associated with enhanced A375 cell line (vemurafenib IC50: 102 nmol/L), sensitivity of cancer cells to CDK4/6 inhibitors (30, 31). A375RV1 and A375RV2 cell lines maintained resistance Interestingly, analysis of whole-exome sequencing data to vemurafenib with IC50 of 1,520 and 1,095 nmol/L, derived from parental A375 and vemurafenib-resistant respectively (Fig. 4A and Table 1). We further charac- A375R1 cells revealed a copy number gain of 11q13 region terized these cells and found that phospho-MEK and in the A375R1 cells and amplification of genes in the 11q13 phospho-ERK levels remained high in the resistant cells locus including CCND1, FGF3, FGF4, and FGF19 (Supple- inthepresenceofvemurafenib concentrations as high mentary Table S1). Copy number variations of CCND1 are as 3 mmol/L compared with parental A375 cells where

2258 Mol Cancer Ther; 13(10) October 2014 Molecular Cancer Therapeutics

Target Vemurafenib Resistance by the CDK4/6 Inhibitor LY2835219

A 120 A375 100 A375-RV1 80 A375-RV2 60 units 40

Figure 4. Cell lines derived 20 from vemurafenib-resistant 0 Relative luminescence tumors maintain resistance to 0 12 3 4 5 vemurafenib. A, sensitivity of A375 Log concentration and tumor-derived resistant A375RV1 and A375RV2 cells to vemurafenib (nmol/L) vemurafenib. Cell viability was assessed using CellTiter Glo. B, B A375 A375-RV1 cyclin D1 elevation and MAPK pathway reactivation in tumor- 0 0.1 0.5 1 5 10 0 0.1 0.5 1 5 10 Vemurafenib (µmol/L) derived A375RV1 cells in the presence of vemurafenib. Cells BRAF were treated with indicated concentrations of vemurafenib ←p61-Splice variant for 24 hours. Cell lysates were analyzed by immunoblotting pMEK1/2 using indicated antibodies.

pERK1/2

Cyclin D1

pRb S780

Tubulin

phospho-MEK and p-ERK were diminished at concen- mechanism of antiproliferative effects of LY2835219 in trations as low as 500 nmol/L (Fig. 4B and Supplemen- parental and vemurafenib-resistant cells. As expected, tary Fig. S3). In addition, we also detected expression of 92% of parental A375 cells arrested in the G1 phase upon p61-B-RAF splice variant in the A375RV1 cells (Fig. 4B), LY2835219 treatment (Fig. 5B). However, LY2835219 suggesting that one mechanism of resistance to vemur- treatment induced cell death in the resistant cells, up afenib in this cell line is likely due to alterations in B- to 90% in the A375RV1 and 69% in A375RV2 cells, in 48 RAF splicing. Consistent with the observations from the hours as indicated by the presence of a subdiploid peak analysis of the tumor lysates, the cell lines derived from in the cytometry histograms. Similarly, LY2835219 the resistant tumors also showed enhanced expression arrested 75% of parental M14 cells in the G1 stage but of cyclin D1 (Fig. 4B and Supplementary Fig. S4C). induced cell death in 70% of vemurafenib-resistant M14R cells (Supplementary Fig. S4A). Thus, antitumor LY2835219 induces G1 arrest in the parental cells but effects of LY2835219 are differentially mediated in apoptosis in vemurafenib-resistant cells derived parental and resistant cells, although the IC50 values from tumors are similar in CellTiter Glo proliferation assay. We We tested whether the vemurafenib-resistant cells further investigated whether LY2835219-induced cell derived from tumors were sensitive to LY2835219 in death in the resistant cells is mediated by apoptosis. cell culture and found that LY2835219 inhibited the As demonstrated in Figs. 5C and 6A, LY2835219 treat- proliferation of the parental A375 and resistant ment signiﬁcantly elevated caspase-3/7 activity in A375RV1 and A375RV2 cells with similar potencies A375RV1 and A375RV2 cells and cleaved PARP activity with IC50 values of 395, 260, and 463 nmol/L, respec- in A375RV1 cells in a concentration-dependent fashion. tively(Fig.5A).Thisisconsistentwiththeresponsesof Similarly, LY2835219 induced higher caspase-3/7 activ- other in vitro vemurafenib-resistant cells. It is well ity and PARP cleavage in other vemurafenib-resistant established that inhibition of cyclin D1–CDK4/6 axis, A375R1 and M14R cells in a concentration-dependent and subsequent inactivation of Rb, arrests proliferating manner (Supplementary Fig. S4B), suggesting that cells in G1 stage of cell cycle. We performed PI staining LY2835219 induces apoptosis preferably in the vemur- and FACS analysis of cell-cycle distribution to study the afenib-resistant cells.

www.aacrjournals.org Mol Cancer Ther; 13(10) October 2014 2259

Yadav et al.

indicating inhibition of CDK4/6 activity and inactivation A 120 of Rb function. Furthermore, LY2835219 treatment induced a concentration-dependent enhancement of 100 A375 80 cyclin D1/D2 proteins in A375 cells, indicative of growth A375-RV1 in vivo 60 A375-RV2 arrest in these cells (Fig. 6A). Consistent with the 40 data, cyclin D1 levels remained high in vemurafenib- 20 resistant tumor cells (Fig. 6A and B and Supplementary Fig. S4C). We explored whether vemurafenib-resistant

Relative luminiscence 0 –1 0 1 2 3 4 5 cells are dependent on cyclin D1 for their survival. We Log concentration LY2835219 (nmol/L) performed shRNA-mediated selective knockdown of B cyclin D1 in the parental and vemurafenib-resistant cells %G1= 69.39 %G1= 92.57 %G1= 82.81 and found that the loss of cyclin D1 decreased phospho- %S = 24.04 %S = 3.70 %S = 8.31 %G –M = 6.58 %G –M = 3.73 %G –M = 8.88 2 2 2 Rb levels in both A375 and A375RV1 cells but induced significantly higher levels of cleaved PARP and cleaved A375 caspase-3 fragments in A375RV1 cells (Fig. 6B). Similarly, specific knockdown of cyclin D1 via siRNA also induced %G1= 54.68 %G1= 53.07 %G1= 42.48 %S = 31.39 %S = 17.33 %S = 39.67 PARP cleavage in A375RV1 and other vemurafenib-resis- %G2–M = 13.92 %G2–M = 29.60 %G2–M = 17.84 %Debris = 90.72 %Debris = 88.77 tant cells including A375RV2 and A375R1 cells (Supple- mentary Fig. S4C), indicating that loss of cyclin D1 induces apoptosis in vemurafenib-resistant cells and not in the %G = 58.08 %G = 58.33 1 1 %G1= 46.14 parental cells. Cyclin D1 knockdown did not affect phos- %S = 30.76 %S = 20.20 %S = 26.95 %G2–M = 11.16 %G2–M = 21.47 %G2–M = 26.91 %Debris = 37.41 %Debris = 69.93 pho-ERK levels (Fig. 6B) or levels of other D-type cyclins, such as cyclin D2 (Supplementary Fig. S4C). Thus, we A375 RV2 A375 RV1 show that vemurafenib-resistant cells are dependent on cyclin D1 for their survival. DMSO 1.5 µmol/L 5 µmol/L LY2835219 Discussion C 600 A375 In this study, we generated multiple in vitro cell lines A375-RV1 and an in vivo model resistant to vemurafenib and 500 A375-RV2 discovered that MAPK reactivation and cyclin D1 ele- 400 vation are associated with acquired resistance in these models. We further describe that inhibition of cyclin 300 D1–CDK4/6 signaling by CDK4/6 inhibitor LY2835219 is an effective therapy to overcome resistance. Expres- 200 sion of B-RAF V600E splice variants, RTK/Ras activa- 100 tion, NRas mutation, and B-RAF amplification are the predominant clinical mechanisms of resistance to Relative caspase-3 activity (%) 0 vemurafenib that have been observed to date, and all 0.0 2.6 3.0 3.5 4.0 of these resistant mechanisms together with B-RAF Log concentration LY2835219 (nmol/L) mutation induce hyperactivation of the MAPK pathway (7, 12, 13). We found that the cells resistant to vemur- Figure 5. CDK4/6 inhibitor LY2835219 is active against and induces afenib with hyperactivation of the MAPK pathway have apoptosis in tumor-derived and vemurafenib-resistant cells. A, sensitivity elevated cyclin D1 expression and that they were sen- of A375, A375RV1, and A375RV2 cells to LY2835219. Cell viability sitive to CDK4/6 inhibitor LY2835219. We subsequently was assessed using CellTiter Glo. B, cell-cycle analysis of A375 and in vivo A375RV1 cells treated by LY2835219. Cells were treated with indicated developed a vemurafenib-resistant model and concentration of LY2835219 for 48 hours and subjected to PI staining and evaluated the antitumor effect of LY2835219 in this subsequent FACS analysis of cell-cycle distribution. Dead cells are model. Consistent with the in vitro findings, phospho- indicated as debris. C, caspase-3/7 activities in LY2835219-treated ERK and cyclin D1 were elevated in these resistant A375, A375RV1, and A375RV2 cells. Cells were treated with increasing concentrations (0, 0.3, 1, 3.3, and 5 mmol/L) of LY2835219 for 24 hours, tumors (Fig. 3B). More importantly, LY2835219 induced and caspase-3/7 activity was determined using Caspase-Glo-3/7 Assay. regression of these vemurafenib-resistant tumors (Fig. 3A). These results suggest that CDK4/6 inhibition represents an effective therapeutic strategy to overcome Upregulation of cyclin D1 is associated with vemurafenib resistance due to MAPK reactivation and LY2835219-induced apoptosis and important for the associated with cyclin D1 elevation. survival of vemurafenib-resistant tumor cells To investigate the mechanism of LY2835219 sensitivity Treatment with LY2835219 caused a concentration- and vemurafenib resistance, we generated resistant cells dependent inhibition of phospho-Rb (S780, S807/S811) lines from vemurafenib-resistant xenograft tumors. with similar potencies in parental and resistant cells, Molecular analysis of the tumor-derived resistant cell

2260 Mol Cancer Ther; 13(10) October 2014 Molecular Cancer Therapeutics

Target Vemurafenib Resistance by the CDK4/6 Inhibitor LY2835219

Figure 6. Cyclin D1 and CDK4/6 signaling is required for the survival A B of vemurafenib-resistant B-RAF A375 A375-RV1 V600E melanoma cells. A, cell 0 0.1 0.5 1 5 10 0 0.1 0.5 1 5 10 LY2835219 (µmol/L) signaling analysis in A375 and pMEK A375-RV1 A375RV1 cells treated with A375 A375-RV1 A375 CDK4/6 inhibitor LY2835219. Cells pERK ++ Control ShRNA were treated with indicated Cyclin D1 Cyclin D1 ShRNA concentrations of LY2835219 for ++ 48 hours. Cell lysates were Cyclin D2 pERK analyzed by immunoblotting using antibodies indicated. B, cyclin D1 pRb S780 Cyclin D1 knockdown by shRNA induced apoptosis in vemurafenib-resistant pRb S807/S811 pRb cells. Cells were transduced with lentivirus encoding either control or p27 cyclin D1 shRNA. Cell lysates were Cleaved PARP collected 72 hours postinfection Cleaved PARP Cleaved- and analyzed by immunoblotting caspase-3 using indicated antibodies. Cell CDK4 Tubulin lysates were collected 72 hours CDK6 posttransfection and analyzed by immunoblotting using indicated Tubulin antibodies.

lines revealed hyperactivation of the MAPK pathway and nation of these 2 agents in delaying and overcoming an increase in cyclin D1 expression just as had been resistance. Additional studies in defining these dose sche- observed in the in vitro resistant cells. Similar to the dules are ongoing in preclinical models. xenograft studies in mice, resistant cells derived from Previous studies revealed that cyclin D1–overexpres- tumors retained resistance to vemurafenib, as well as sing cells demonstrate constitutive CDK activity and sensitivity to LY2835219 (Figs. 4A and 5A). We further increased sensitivity to CDK4/6 inhibitors (30, 31). Cyclin examined the mechanism of antiproliferative effects of D1 was also demonstrated to be implicated in resistance to LY2835219 in parental and resistant cells. As expected, inhibitors of ERBB2, EGFR, and ER signaling (32, 33). LY2835219 arrested parental cells in the G1 stage (Fig. 5B Recent studies suggest that cyclin D1 overexpression may and Supplementary Fig. S4A), consistent with the role of be sufficient to render B-RAF V600E melanoma cells CCND1 CDK4/6 function in progression of cells from G1 to S resistant to B-RAF inhibition (28). is amplified in stage of cell cycle (25). However, treatment of resistant 11% of melanoma (2), including 17% of B-RAF V600E cells with LY2835219 induced apoptosis within 24 to 48 melanoma (28), indicating that it could be a mechanism hours in a concentration-dependent fashion (Figs. 5B of de novo resistance to B-RAF inhibitors. Molecular anal- and C and 6A and Supplementary Fig. S4A and S4B). ysis of our vemurafenib-resistant tumors revealed signif- The surprising pro-apoptotic effects of LY2835219 were icant upregulation of cyclin D1 levels relative to parental observed across a variety of vemurafenib-resistant cell tumors (Fig. 3B). Consistently, cyclin D1 was also upre- lines, including cells derived from resistant xenograft gulated across most vemurafenib-resistant cell lines, tumors. Thus, vemurafenib-resistant B-RAF V600E mel- including cell lines derived from resistant xenograft anomas with hyperactivated MAPK signaling and tumors. The cyclin D1 upregulation could simply results enhanced cyclin D1 expression are prone to apoptosis from MAPK reactivation, as it is a downstream effector of upon CDK4/6 inhibition by LY2835219, suggesting that MAPK signaling. To date, we have not fully characterized these resistant cells are more dependent on cyclin D1– the cyclin D1 amplification in every resistant cell line. CDK4/6 signaling for survival. However, in at least A375R1 case, we found that the Both vemurafenib and LY2835219 were demonstrated vemurafenib resistance was associated with 11q13 copy to inhibit tumor growth of B-RAF V600E melanoma in number gain (Supplementary Table S1). This chromosom- single agent, and combination of these 2 resulted in an al region includes the genes for FGF ligands as well as additive tumor growth inhibition (Fig. 2). This suggests CCND1, potentially explaining both the cyclin D1 upre- that upfront combination of CDK4/6 and B-RAF inhibi- gulation and CDK4/6 dependence described here, as well tors may be more efficacious than single-agent therapy. as the previously reported FGF pathway activation (10). However, a more robust tumor growth regression was Analysis of cyclin D1 amplification in other resistant cells observed when CDK4/6 inhibitor LY2835219 was used is ongoing. We further investigated whether cyclin D1 is for the treatment of xenograft tumors acquired resistance required for proliferation and survival of vemurafenib- to vemurafenib (Fig. 3A). These results suggest that resistant cells. Consistent with the pro-apoptotic effects of vemurafenib followed by LY2835219 treatment schedule LY2835219, knockdown of cyclin D1 by shRNA also might be a more effective approach than upfront combi- induced higher levels of apoptosis in vemurafenib-

www.aacrjournals.org Mol Cancer Ther; 13(10) October 2014 2261

Yadav et al.

resistant versus parental cells (Fig. 6B and Supplementary Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): V. Yadav, T.F. Burke, L. Huber, R.D. Van Horn, Fig. S4C). However, these data do not entirely exclude the Y. Zhang possibility that either off-target effects of LY2835219 or Analysis and interpretation of data (e.g., statistical analysis, biostatis- tics, computational analysis): V. Yadav, T.F. Burke, Y. Zhang, S.G. Bucha- inhibition of Rb-independent pathways by cyclin D1 nan, E.M. Chan, R.P. Beckmann, S.-B. Peng knockdown might also contribute to the induction of Writing, review, and/or revision of the manuscript: V. Yadav, T.F. Burke, apoptosis observed selectively in vemurafenib-resistant S.G. Buchanan, E.M. Chan, J.J. Starling, R.P. Beckmann, S.-B. Peng Administrative, technical, or material support (i.e., reporting or orga- cells. Altogether, we demonstrate that cyclin D1 is impor- nizing data, constructing databases): V. Yadav, S.-B. Peng tant for survival of vemurafenib-resistant cells with Study supervision: V. Yadav, E.M. Chan, R.P. Beckmann, S.-B. Peng hyperactivation of the MAPK pathway and cyclin D1 upregulation and propose LY2835219, a CDK4/6 dual Acknowledgments inhibitor, as a potential therapy to overcome such The authors thank Dr. Philip J. Elbert for DNA sequence and mutational resistance. analysis of A375- and A375R1-resistant cell lines.

Disclosure of Potential Conﬂicts of Interest T.F. Burke, L. Huber, E.M. Chan, and R.P. Beckmann have ownership Grant Support interest (including patents) in Eli Lilly and Co. No potential conﬂicts of This work was supported by Eli Lilly and Company. interest were disclosed by the other authors. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked Authors' Contributions advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate Conception and design: V. Yadav, S.G. Buchanan, E.M. Chan, this fact. R.P. Beckmann, S.-B. Peng Development of methodology: V. Yadav, T.F. Burke, L. Huber, R.D. Van Received March 25, 2014; revised July 30, 2014; accepted August 1, 2014; Horn, Y. Zhang, E.M. Chan, R.P. Beckmann, S.-B. Peng published OnlineFirst August 13, 2014.

References 1. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, et al. mediated acquired B-RAF inhibitor resistance. Nat Commun 2012; Mutations of the BRAF gene in human cancer. Nature 2002;417: 3:724. 949–54. 14. Das Thakur M, Salangsang F, Landman AS, Sellers WR, Pryer NK, 2. Hodis E, Watson IR, Kryukov GV, Arold ST, Imielinski M, Theurillat JP, Levesque MP, et al. Modelling vemurafenib resistance in melanoma et al. A landscape of driver mutations in melanoma. Cell 2012;150: reveals a strategy to forestall drug resistance. Nature 2013;494: 251–63. 251–5. 3. Bollag G, Hirth P, Tsai J, Zhang J, Ibrahim PN, Cho H, et al. Clinical 15. Johannessen CM, Boehm JS, Kim SY, Thomas SR, Wardwell L, efficacy of a RAF inhibitor needs broad target blockade in BRAF- Johnson LA, et al. COT drives resistance to RAF inhibition through mutant melanoma. Nature 2010;467:596–9. MAP kinase pathway reactivation. Nature 2010;468:968–72. 4. Chapman PB, Hauschild A, Robert C, Haanen JB, Ascierto P, Larkin J, 16. Montagut C, Sharma SV, Shioda T, McDermott U, Ulman M, Ulkus LE, et al. Improved survival with vemurafenib in melanoma with BRAF et al. Elevated CRAF as a potential mechanism of acquired resistance V600E mutation. N Engl J Med 2011;364:2507–16. to BRAF inhibition in melanoma. Cancer Res 2008;68:4853–61. 5. Flaherty KT, Puzanov I, Kim KB, Ribas A, McArthur GA, Sosman JA, 17. Paraiso KH, Xiang Y, Rebecca VW, Abel EV, Chen YA, Munko AC, et al. et al. Inhibition of mutated, activated BRAF in metastatic melanoma. PTEN loss confers BRAF inhibitor resistance to melanoma cells N Engl J Med 2010;363:809–19. through the suppression of BIM expression. Cancer Res 2011;71: 6. Long GV, Trefzer U, Davies MA, Kefford RF, Ascierto PA, Chapman PB, 2750–60. et al. Dabrafenib in patients with Val600Glu or Val600Lys BRAF-mutant 18. Villanueva J, Vultur A, Lee JT, Somasundaram R, Fukunaga-Kalabis M, melanoma metastatic to the brain (BREAK-MB): a multicentre, open- Cipolla AK, et al. Acquired resistance to BRAF inhibitors mediated by a label, phase 2 trial. Lancet Oncol 2012;13:1087–95. RAF kinase switch in melanoma can be overcome by cotargeting MEK 7. Nazarian R, Shi H, Wang Q, Kong X, Koya RC, Lee H, et al. Melanomas and IGF-1R/PI3K. Cancer Cell 2010;18:683–95. acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS 19. Shi H, Kong X, Ribas A, Lo RS. Combinatorial treatments that over- upregulation. Nature 2010;468:973–7. come PDGFRbeta-driven resistance of melanoma cells to V600EB- 8. Su F, Bradley WD, Wang Q, Yang H, Xu L, Higgins B, et al. Resistance RAF inhibition. Cancer Res 2011;71:5067–74. to selective BRAF inhibition can be mediated by modest upstream 20. Olson OC, Joyce JA. Microenvironment-mediated resistance to anti- pathway activation. Cancer Res 2012;72:969–78. cancer therapies. Cell Res 2013;23:179–81. 9. Wagle N, Emery C, Berger MF, Davis MJ, Sawyer A, Pochanard P, et al. 21. Wilson TR, Fridlyand J, Yan Y, Penuel E, Burton L, Chan E, et al. Dissecting therapeutic resistance to RAF inhibition in melanoma by Widespread potential for growth-factor-driven resistance to antican- tumor genomic profiling. J Clin Oncol 2011;29:3085–96. cer kinase inhibitors. Nature 2012;487:505–9. 10. Yadav V, Zhang X, Liu J, Estrem S, Li S, Gong XQ, et al. Reactivation of 22. Kim KB, Kefford R, Pavlick AC, Infante JR, Ribas A, Sosman JA, et al. mitogen-activated protein kinase (MAPK) pathway by FGF receptor 3 Phase II study of the MEK1/MEK2 inhibitor Trametinib in patients with (FGFR3)/Ras mediates resistance to vemurafenib in human B-RAF metastatic BRAF-mutant cutaneous melanoma previously treated with V600E mutant melanoma. J Biol Chem 2012;287:28087–98. or without a BRAF inhibitor. J Clin Oncol 2013;31:482–9. 11. Straussman R, Morikawa T, Shee K, Barzily-Rokni M, Qian ZR, Du J, 23. Hacker E, Muller HK, Irwin N, Gabrielli B, Lincoln D, Pavey S, et al. et al. Tumour micro-environment elicits innate resistance to RAF Spontaneous and UV radiation-induced multiple metastatic mela- inhibitors through HGF secretion. Nature 2012;487:500–4. nomas in Cdk4R24C/R24C/TPras mice. Cancer Res 2006;66: 12. Poulikakos PI, Persaud Y, Janakiraman M, Kong X, Ng C, Moriceau G, 2946–52. et al. RAF inhibitor resistance is mediated by dimerization of aberrantly 24. Hayward NK. Genetics of melanoma predisposition. Oncogene 2003; spliced BRAF(V600E). Nature 2011;480:387–90. 22:3053–62. 13. Shi H, Moriceau G, Kong X, Lee MK, Lee H, Koya RC, et al. Melanoma 25. Malumbres M, Barbacid M. To cycle or not to cycle: a critical decision in whole-exome sequencing identifies (V600E)B-RAF amplification- cancer. Nat Rev Cancer 2001;1:222–31.

2262 Mol Cancer Ther; 13(10) October 2014 Molecular Cancer Therapeutics

Target Vemurafenib Resistance by the CDK4/6 Inhibitor LY2835219

26. Smalley KS, Haass NK, Brafford PA, Lioni M, Flaherty KT, Herlyn M. activities alone/in combination with gemcitabine. Invest New Drugs. Multiple signaling pathways must be targeted to overcome drug 2014 Jun 13. [Epub ahead of print]. resistance in cell lines derived from melanoma metastases. Mol Cancer 30. Finn RS, Dering J, Conklin D, Kalous O, Cohen DJ, Desai AJ, et al. PD Ther 2006;5:1136–44. 0332991, a selective cyclin D kinase 4/6 inhibitor, preferentially inhibits 27. Bhatt KV, Spofford LS, Aram G, McMullen M, Pumiglia K, Aplin AE. proliferation of luminal estrogen receptor-positive human breast can- Adhesion control of cyclin D1 and p27Kip1 levels is deregulated in cer cell lines in vitro. Breast Cancer Res 2009;11:R77. melanoma cells through BRAF-MEK-ERK signaling. Oncogene 2005; 31. Lapenna S, Giordano A. Cell cycle kinases as therapeutic targets for 24:3459–71. cancer. Nat Rev Drug Discov 2009;8:547–66. 28. Smalley KS, Lioni M, Dalla Palma M, Xiao M, Desai B, Egyhazi S, et al. 32. Musgrove EA, Sutherland RL. Biological determinants of endocrine Increased cyclin D1 expression can mediate BRAF inhibitor resistance resistance in breast cancer. Nat Rev Cancer 2009;9:631–43. in BRAF V600E-mutated melanomas. Mol Cancer Ther 2008;7: 33. Kalish LH, Kwong RA, Cole IE, Gallagher RM, Sutherland RL, Mus- 2876–83. grove EA. Deregulated cyclin D1 expression is associated with 29. Gelbert LM, Cai S, Lin X, Sanchez-Martinez C, Del Prado M, Lallena decreased efﬁcacy of the selective epidermal growth factor receptor MJ, et al. Preclinical characterization of the CDK4/6 inhibitor tyrosine kinase inhibitor geﬁtinib in head and neck squamous cell LY2835219: in-vivo cell cycle-dependent/independent anti-tumor carcinoma cell lines. Clin Cancer Res 2004;10:7764–74.

www.aacrjournals.org Mol Cancer Ther; 13(10) October 2014 2263

The CDK4/6 Inhibitor LY2835219 Overcomes Vemurafenib Resistance Resulting from MAPK Reactivation and Cyclin D1 Upregulation

Vipin Yadav, Teresa F. Burke, Lysiane Huber, et al.

Mol Cancer Ther 2014;13:2253-2263. Published OnlineFirst August 13, 2014.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-14-0257

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2014/08/16/1535-7163.MCT-14-0257.DC1

Cited articles This article cites 32 articles, 11 of which you can access for free at: http://mct.aacrjournals.org/content/13/10/2253.full#ref-list-1

Citing articles This article has been cited by 13 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/13/10/2253.full#related-urls

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 Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/13/10/2253. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.