Exome Sequencing of Plasma DNA Portrays the Mutation Landscape Of

Published OnlineFirst March 27, 2018; DOI: 10.1158/1078-0432.CCR-18-0103

Precision Medicine and Imaging Clinical Cancer Research Exome Sequencing of Plasma DNA Portrays the Mutation Landscape of Colorectal Cancer and Discovers Mutated VEGFR2 Receptors as Modulators of Antiangiogenic Therapies Rodrigo A. Toledo1,2, Elena Garralda2,3,4, Maria Mitsi5, Tirso Pons6, Jorge Monsech7, Estela Vega3,4, Alvaro Otero8, Maria I. Albarran9, Natalia Banos~ 1, Yolanda Duran 1, Victoria Bonilla1, Francesca Sarno1, Marta Camacho-Artacho10, Tania Sanchez-Perez11, Soﬁa Perea1, Rafael Alvarez 3,4, Alba De Martino12, Daniel Lietha10, Carmen Blanco-Aparicio9, Antonio Cubillo3,4, Orlando Domínguez7, Jorge L. Martínez-Torrecuadrada8, and Manuel Hidalgo13

Abstract

Purpose: Despite the wide use of antiangiogenic drugs in 1% to 3% of samples from cancer sequencing projects harbor the clinical setting, predictive biomarkers of response to these KDR somatic mutations located in protein residues frequently drugs are still unknown. mutated in other cancer-relevant kinases, such as EGFR, ABL1, Experimental Design: We applied whole-exome sequenc- and ALK. Our in vitro and in vivo functional assays conﬁrmed that ing of matched germline and basal plasma cell-free DNA L840F causes strong resistance to antiangiogenic drugs, whereas samples (WES-cfDNA) on a RAS/BRAF/PIK3CA wild-type the KDR hot-spot mutant R1032Q confers sensitivity to strong metastatic colorectal cancer patient with primary resistance VEGFR2 inhibitors. Moreover, we showed that the D717V, to standard treatment regimens, including inhibitors to the G800D, G800R, L840F, G843D, S925F, R1022Q, R1032Q, and VEGF:VEGFR2 pathway. We performed extensive functional S1100F VEGFR2 mutants promote tumor growth in mice. experiments, including ectopic expression of VEGFR2 Conclusions: Our study supports WES-cfDNA as a power- mutants in different cell lines, kinase and drug sensitivity ful platform for portraying the somatic mutation landscape of assays, and cell- and patient-derived xenografts. cancer and discovery of new resistance mechanisms to cancer Results: WES-cfDNA yielded a 77% concordance rate with therapies. Importantly, we discovered that VEGFR2 is somat- tumor exome sequencing and enabled the identiﬁcation of the ically mutated across tumor types and that VEGFR2 mutants KDR/VEGFR2 L840F clonal, somatic mutation as the cause of can be oncogenic and control sensitivity/resistance to anti- therapy refractoriness in our patient. In addition, we found that angiogenic drugs. Clin Cancer Res; 24(15); 3550–9. 2018 AACR.

Introduction cizumab, as well as to new agents targeting VEGFR2, such as regorafenib, is still a fundamental unmet medical necessity (4). Colorectal cancer is currently the third most frequent cancer Genetic variants of the VEGF:VEGFR1/2 pathway could influ- diagnosed worldwide and is predicted to reach 2.2 million new ence the outcome of antiangiogenic treatment. Although some cases per year by 2030 (1). Although bevacizumab treatment studies have suggested the potential association of tumor (anti-VEGF) increases the overall survival of patients with meta- response with VEGF/VEGFR germline polymorphisms (5), these static colorectal cancer (mCRC), approximately 50% and 80% of results could not be confirmed in subsequent assessments (6). patients in first and second lines of treatment respectively are Interestingly, recent in vitro studies demonstrated that VEGFR2 refractory and do not benefit from this therapy strategy (2, 3). The plays a prominent role not only in endothelial cells, as is usually identification of predictive biomarkers of the response to beva- assumed, but also in cancer cells (7). However, it is largely

1Gastrointestinal Cancer Clinical Research Unit, Spanish National Cancer Note: Supplementary data for this article are available at Clinical Cancer Research Center (CNIO), Madrid, Spain. 2Vall d'Hebron Institute of Oncology Research Online (http://clincancerres.aacrjournals.org/). (VHIO), CIBERONC, Barcelona, Spain. 3Centro Integral Oncologico Clara Campal 4 Current address for T. Pons: Stem Cells and Immunity Lab, Department of (CIOCC), Hospital Universitario HM Sanchinarro, Madrid, Spain. Universidad San Immunology and Oncology, National Centre for Biotechnology, Madrid, Spain. Pablo CEU, Madrid, Spain. 5Laboratory of Biomolecular Research, Paul Scherrer Current address for M. Mitsi: Institute of Food Nutrition and Health, ETH Zurich, Institute, Villigen, Switzerland. 6Structural Computational Biology Group, CNIO, Switzerland. Madrid, Spain. 7Genomics Core Unit, CNIO, Madrid, Spain. 8Crystallography and Protein Engineering Unit, CNIO, Madrid, Spain. 9Experimental Therapeutics Corresponding Author: Rodrigo A. Toledo, Vall d'Hebron Institute of Oncology Programme, CNIO, Madrid, Spain. 10Cell Signalling and Adhesion Group, CNIO, (VHIO), CIBERONC, Barcelona 08035, Spain. Phone: 34-932543450; E-mail: Madrid, Spain. 11Molecular Genetics of Angiogenesis Laboratory, Spanish Nation- [email protected] 12 al Center for Cardiovascular Research (CNIC), Madrid, Spain. Histopathology doi: 10.1158/1078-0432.CCR-18-0103 Unit, CNIO, Madrid, Spain. 13Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts. 2018 American Association for Cancer Research.

3550 Clin Cancer Res; 24(15) August 1, 2018

Exome Sequencing of Plasma DNA and VEGFR2 Somatic Mutations

Plasma- and DNAseq Kits (Rubicon Genomics Inc.), respectively. Translational Relevance Barcode indices were added to samples during eight PCR cycles of Our work illustrates the high capacity of whole-exome se- template preparation, and 550 ng of each sample was processed quencing of plasma cfDNA (WES-cfDNA) in portraying the through the SureSelectXT Target Enrichment System (Agilent somatic mutation landscape in cancer and its potential as a SureSelect V5, ref. 5190-6208, protocol G7530-90000 version global tumor-free genomic platform to explore the genetic B1). xGen Blocking Oligos (IDT) were used as suggested by causes of primary resistance to cancer therapies. The implemen- Rubicon Genomics. Captured targets were subsequently enriched tation of WES-cfDNA would enable discovery studies on cancer by 11 cycles of PCR with KAPA HiFi HotStart (Kapa Biosystems), patients for whom only blood/plasma samples are available, with a Tann of 60 and the following primers, which target generic as well as serve as a complementary assay to the standard ends of Illumina adapters: AATGATACGGCGACCACCGAGAT WES-tumor by increasing the overall capacity for identifying and CAAGCAGAAGACGGCATACGAGAT. For sequencing, mag- somatic mutations and molecular targets. As a proof of concept, netic bead–purified libraries with similar concentrations of WES-cfDNA identified the L840F VEGFR2 clonal, somatic muta- cfDNAs and tDNA, and half the concentration of gDNA were tion in a highly refractory metastatic colorectal cancer patient. pooled in order to increase coverage and favor the detection of Our recognition of KDR/VEGFR2 somatic mutations, occurring noninherited subclonal mutations. Sequencing was carried out in in 1% to 3% of human cancers, as oncogenic and capable of the Illumina HiSeq4000 platform. All sequencing data are going modulating the efficacy of antiangiogenic cancer therapies to be deposited in the European Nucleotide Archive (ENA) under in vitro and in vivo will prompt future investigation to define the accession number ENA#202177, at the time of publication. the potential impact of these mutants in the clinical setting. Somatic mutation call Bioinformatics analyses were performed using the NEXTGEN software (Softgenetics) (9). FastaQ files were aligned using the unknown whether clinically relevant KDR/VEGFR2 mutations BWA pipeline, and the variants were processed by sequential occur in tumor cells and potentially regulate drug efficiency. stringent filters to exclude low-confidence variants. Only variants that passed the following filters were classified as high-quality and Materials and Methods considered in the study: overall and allele scores 12; coverage Study supervision 20; number of mutated reads 20; percentage of mutated reads The study was approved by the Institutional Review Boards of 3% of cfDNA/tDNA and 35% of gDNA; F:R read balance 0.1; Hospital Universitario HM Sanchinarro and conducted in agree- and F:R read percentage 0.45. The list of nonhereditary muta- ment with the Declaration of Helsinki and the International tions detected by WES-cfDNA and WES-tumor was generated after Conference on Harmonization of Good Clinical Practice guide- disregarding germline variants (obtained by WES-gDNA). A lines. The patient gave written informed consent to participate in detailed genomic annotation of the somatic mutations we iden- the study. Mice used in this research were treated humanely tified, prediction of mutation pathogenicity based on predictor according to the regulations laid down by the Spanish National algorithms (SIFT, Polyphen2, LRT, Mutation Taster, Mutation Cancer Research Centre (CNIO) Bioethics Committee. Assessor, and other software packages), allele frequencies in population studies, such as 1000G and EXAC, and additional DNA extraction information are shown in Supplementary Table S1. DNA was extracted from leukocytes (gDNA), liver metastasis (tDNA), and basal and on-treatment plasma samples (cfDNA), TaqMan SNP genotyping assay using commercial kits according to the manufacturer's instruc- A custom TaqMan genotyping assay for the detection of the tions (Qiagen). The DNA amount was quantified with a Qubit KDR c.2518C (L840L) and KDR c.2518C>T (L840F) alleles was Fluorometer (Thermo Fisher) and reported in nanogram. cfDNA designed using the Thermo Fisher online Design Tool (oligonu- samples were also quantified using a modified version of human cleotides and probes are shown in Supplementary Table S2). LINE-1–based quantitative real-time PCR and reported in genome equivalents (GE; GE being one haploid human genome weighing Genetic/protein database and protein structure analyses 3.3 pg). gDNA and tDNA were sheared to 300-bp fragments on a Previously reported germline and somatic variants in KDR were Covaris instrument (Covaris) according to standard procedures. retrieved from general population (EXACT and ESP) and cancer The 2100 Bioanalyzer (Agilent) was used to access the quality and (COSMIC, GENIE, and PCAWGS) sequencing public projects. The size of the preprocessed and postprocessed samples and libraries. VEGFR2, EGFR, and ABL1 protein structures were obtained from the RCSB data bank; structurally analogous mutations in other cancer- Routine genetic analysis relevant kinases were identified using MutationAligner; kinase The FDA-approved Cobas mutation Kit (Roche) was used to residues interacting with kinase inhibitors were mapped using the analyze the following mutations in the diagnostic biopsy tDNA: LigPlot software. Computational modeling of inhibitor binding to KRAS (G12S/R/C/V/A/D, G13D, Q61H, A146T), NRAS (Q61K/R/ WT and L840F VEGFR2 was performed as previously described. L/H), BRAF (V600E), and PIK3CA (E542K, E545K/G, Q546K, M1043I, and H1047Y/R/L). The presence of the same mutations Generation and treatment of the Avatar patient-derived in the patient's basal and on-treatment cfDNA samples was xenograft model assessed by the highly sensitive BEAMing technique (8). Liver metastasis biopsy was performed after tumor progression to capecitabine–bevacizumab rechallenge (Fig. 1). A fraction of Whole-exome sequencing the biopsy was used to generate the Avatar model as previously Sequencing libraries of cfDNA (15 ng), and gDNA and described by our group (10). Expanded cohorts (five to six tDNA (70–110 ng) samples were prepared using the ThruPLEX animals per arm) were treated with anti-VEGF drugs (B20/murine

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3551

Toledo et al.

gDNA A Diagnosc biopsy plasma (cfDNA) Plasma (cfDNA)

Baseline PD to FOLFIRI-cetuximab PD to FOLFOX-bevacizumab PD to afanib-cetuximab

Biopsy for Avatar and tDNA

PD to afanib-cetuximab PD to capecitabine-bevacizumab PD to oncolyc adenovirus PD to regorafenib

B Liver metastasis diagnosc biopsy Basal plasma cell-free DNA On-treatment liver metastasis biopsy 1 2 3 4 COBAS Gene panel liquid biopsy (BEAMing) Avatar mouse model tDNA

KRAS wt Whole-exome NRAS wt sequencing BRAF wt PIK3CA wt 3a

TaqMan 3b KDR VEGFR2 3c Conﬁrmed by Conﬁrmed by Sanger sequencing genotyping assay p.L840F 3b 3a 3c

KDR L840F cfDNAs KDR L840F tDNA

KDR wt water leukocytes

Coverage: 248 A=28 (11,3%, 9+,19-)

Figure 1. Discovery of the KDR/VEGFR2 L840F somatic mutation in a refractory mCRC patient. A, Thoracic computed tomography scans of the refractory mCRC patient. Imaging examinations show no response to any treatment and disease progression to FOLFIRI-cetuximab, FOLFOX-bevacizumab, afatinib-cetuximab, capecitabine-bevacizumab, oncolytic adenovirus and regorafenib. The time points of sample collection are pointed with colored arrows: red for diagnostic biopsy, blue for DNA sequencing, and green for Avatar generation and tumor DNA sequencing. Metastatic foci are indicated by black arrows. B, Noninvasive strategy for the discovery of primary therapy-resistance gene(s). Genetic and genomic analyses were performed according to the DNA sample availability. First, tumor biopsy was used (and exhausted) on routine genetic analysis that revealed no mutations in the main resistance-associated genes KRAS/NRAS/BRAF/PIK3CA. Second, whole-exome sequencing was performed using basal plasma cfDNA (to obtain both germline and cancer genetic variants/mutations) and matched leukocyte DNA (to filter out germline variants), and identified the KDR/VEGFR2 L840F somatic mutation. Third, the mutation was confirmed by Sanger and TaqMan assays in pre- and postplasma cfDNA samples and in a second biopsy. A fragment of the biopsy was used to generated the patient-derived xenograft (PDX) Avatar model and for tumor DNA exome sequencing.

and bevacizumab/human), VEGFR2 kinase inhibitors (axitinib, the Colo-320 cell line to interrogate the phenotypic changes cabozantinib, cabozantinib:MEK inhibitor combo, lenvatinib, caused by the overexpression of VEGFR2 mutants because it pazopanib, regorafenib, and sorafenib), and inhibitors of other has the same genetic background as the patient's tumor kinases, such as afatinib (EGFR), crizotinib (MET), and MEK (mutated TP53/APC and WT KRAS/BRAF). The MDST8 colo- inhibitor (MAPK). Information on the treatment regimens is rectal cancer cell line was used for drug sensitivity studies shown in Supplementary Table S3. because it naturally harbors the KDR/VEGFR2 R1032Q mutation, which we found to be a hot-spot VEGFR2 mutation in Cell lines human cancers. The human colorectal cancer cell lines used in the current Colo320 and MDST8 colorectal cell lines were obtained from study were selected based on their genotype in order to be as the ATCC and cultured at 37 Cin5%CO2, in RPMI Medium informative as possible for each experiment. Thus, we chose 1640 þ GlutaMAX (Gibco) and DMEM þ 2 mmol/L Glutamine

3552 Clin Cancer Res; 24(15) August 1, 2018 Clinical Cancer Research

Exome Sequencing of Plasma DNA and VEGFR2 Somatic Mutations

(Gibco), respectively, supplemented with 10% FBS (Thermo nal antibody (D9E) to p-AKT (1:300; Cell Signaling Techno- Fisher Scientific). Porcine aortic endothelial (PAE) cell lines, logy; cat. Nr 4060), followed by incubation with Alexa kindly provided by Dr. Kurt Ballmer-Hofer, were grown in DMEM Fluor 555–conjugated donkey anti-rabbit IgG (1:400; Life supplemented with 10% FBS. Technologies cat. Nr A27039) at 37C for 20 minutes. Nuclei were counterstained with DAPI (Molecular Probes) at 1:1,000 Generation of stable colorectal and endothelial cell lines dilution, and the slides were mounted with Mowiol 4-88 Colo-320 and PAE cell lines were used to generate cell lines (Calbiochem). Images were acquired with a confocal TCS-SP5 stably expressing the VEGFR2 mutants. Cells were seeded in (AOBS-UV; Leica Microsystems) confocal microscope, 10-cm plates in the appropriate medium and were grown to equipped with a 20xHCX PL APO 0.7 N.A. objective. 70% confluence. Transfection with constructs carrying either the empty vector (EV) or VEGFR2 (WT or mutant) was performed Proliferation assays with polyethylenimine (PEI). Briefly, 30 mg of WT or mutant Proliferation assays were performed using the CellTiter-Glo VEGFR2 plasmid (in the pBE vector containing the neomycine Luminescent Cell Viability Assay (Promega). Briefly, cell lines resistance gene, which confers resistance to the selection antibiotic were seeded in 96-well microtiter plates at a density of 10,000 G418) was mixed with 60 mL PEI (1 mg/mL in H2O) in 2 mL cells/well and were incubated for 24 hours before adding the serum-free DMEM, incubated for 10 minutes at room temperature various drugs. A "mother plate" containing drugs at a concen- and added to the cells. Following a 3-hour incubation at 37C, the tration 200 higher than the final concentration to be used in medium was changed, and the cells were allowed to grow to 100% the cell culture was prepared by serial dilutions of stock solu- confluence. Cells were reseeded at a series of dilutions (1:1,000– tions of the drugs (10 mmol/L) in DMSO. The appropriate 1:5,000) in antibiotic selection medium (1 mg/mL G418) to volume from each drug (usually 2 mL) was added automatically allow for single colonies to grow, whereas nontransfected cells (Beckman FX 96 tip) from this plate to the cell culture plate to were dying. Individual colonies were consecutively transferred reach the final concentration for each drug. Each concentration to 24-well and 6-well plates and screened by Western blotting was assayed twice. The final concentration of DMSO in the for VEGFR2 expression. To reduce polyclonality, colonies with tissue culture media did not exceed 1%. The cells were exposed the highest expression levels were subjected to three additional to the drugs for 72 hours and then analyzed using the CellTiter- rounds of subcloning. Glo Luminescent Cell Viability Assay (Promega). Cell proliferation values were plotted against drug concentrations and fitted VEGF stimulation and Western blotting to a sigmoid dose-response curve using the Activity base soft- Transiently transfected HEK293 cells or stable PAE cell lines ware from IDBS in order to calculate growth inhibition (GI50) expressing WT or L840F-KDR were starved in DMEM supple- values versus DMSO. mented with 1% BSA for 4 hours at 37Candweresubse- quently stimulated with 1.5 nmol/L (60 ng/mL) VEGF165 for 10 Cloning and mutagenesis minutes at 37C. Following stimulation, the cells were scraped The L840F and eight additional KDR mutations of interest in lysis buffer (50 mmol/L Tris, pH ¼ 8.0, 120 mmol/L NaCl, identified on cancer databanks, as well as the K868M kinase-dead and 1% NP-40) supplemented with protease inhibitors (Roche; mutation, were generated by site-directed mutagenesis of WT cat. Nr 04693159001) and phosphatase inhibitors (1 mmol/L KDR/VEGFR2 cloned on the pBE vector, using the QuikChange sodium orthovanadate and 20 mmol/L phernylarsine oxide) Kit (Agilent) and the primers described in Supplementary Table and incubated for 30 minutes on ice. Cell lysates were collected S4. Mutations were confirmed by Sanger sequencing of the entire as the supernatant of a centrifugation at 30,000 g for 15 open reading frame. minutes and subjected to Western blot analysis. The following antibodies were used to probe receptor activation: total KDR Transfection and xenograft models (Cell Signaling Technology, cat. Nr 2479) and phospho KDR at Colo-320 cell line–derived xenografts were generated from 5 Y1175 (Cell Signaling Technology, cat. Nr 2478). The second- subcutaneous injections of 4 10 cells resuspended in PBS in ary antibodies used were alkaline phosphatase (AP) conjugated 4 nude mice per genotype. Tumors were measured weekly, and the (Southern Biotech). All antibodies were diluted at a 1:1,000 animals were sacrificed within 2 months or when tumors reached ratio in 5% BSA in Tris-buffered saline, containing 0.05% the established humane endpoint. Mice injected with EV or the Tween20 (TBST) buffer. The chemiluminescence signal was K868M kinase-dead mutant were kept alive and monitored week- developed with the Novex AP Chemiluminescence substrate ly for 4 months. (Invitrogen; cat. Nr 100002906), recorded with an Amersham Imager 600 (GE Healthcare), and quantified by ImageJ (NIH). Production of recombinant kinase domains of WT, L840F, and Activation of KDR was assessed by the ratio of phospho-to-total R1032Q VEGFR2 signal. The kinase domains (residues 806–1171) of WT, L840F, and R1032Q VEGFR2 without the kinase insert domain (aa 940– Tissue immunofluorescence 989) were cloned, tagged with 6 His at their C-terminus, and Immunofluorescence staining was performed to detect p-ERK expressed in the baculovirus-infected insect cell system. Pro- and p-AKT. Formalin-fixed and paraffin-embedded tumors teins were purified by affinity chromatography on HisTrap from Avatar models were cut into 3-mm-thick sections, depar- columns, followed by size-exclusion chromatography on a affinized,andpreincubatedwithFBStopreventnonspecific HiLoad 16/600 Superdex 200 prep grade column (GE Health- binding. The sections were incubated at room temperature for care), using an AKTA€ system (GE Healthcare). Fractions con- 30 minutes with a rabbit polyclonal antibody to p-ERK (1:300; taining kinase domains were identified by SDS-PAGE and Cell Signaling Technology; cat. Nr 9101) or a rabbit monoclo- concentrated by ultrafiltration up to 0.2 mg/mL. Protein

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3553

Toledo et al.

mutations were confirmed by in-gel enzymatic digestion fol- Results lowed by LC-MS/MS analysis. Clinical Biochemical assays Case report. A 56-year-old man was diagnosed in our center with The kinase activity of recombinant WT, L840F, and R1032Q mCRC with liver and lung metastases (cT4N2M1). Highly VEGFR2 kinase domains, as well as that of a commercially sensitive Cobas and BEAMing assays were performed, and no fi available recombinant VEGFR2 cytoplasmic domain (residues hot-spots mutations were identi ed in the KRAS/NRAS/ 789–1356; PV3660; Thermo Fisher) that was used as a positive PIK3CA/BRAF genes in the patient's cfDNA, nor in tumor control were analyzed using the LANCE Ultra time-resolved biopsy samples. He received FOLFIRI-cetuximab as frontline fluorescence resonance energy transfer assay from Perkin Elmer treatment, but there was evidence of tumor progression after fi fi according to the manufacturer's instructions. Briefly, the enzymes the rst tumor evaluation. Consequently, he received ve were titrated starting from an initial concentration of 5 mg/mL and additional lines of treatment: FOLFOX-bevacizumab; afatinib proceeding with 1:4 serial dilutions, and were added to the with cetuximab; oncolytic adenovirus monotherapy (the last reaction buffer (15 mmol/L HEPES pH 7.4, 20 mmol/L NaCl, two lines of treatment in the context of phase I clinical trials); rechallenge with capecitabine-bevacizumab; and finally, regor- 1 mmol/L EGTA, 0.02% Tween 20, 10 mmol/L MgCl2, 0.1 mg/mL BGG, 2 mmol/L DTT), containing 15 mmol/L ATP and 200 nmol/L afenib. However, there was no disease stability or response to Ultralight-labeled Poly GT substrate in a total volume of 20 mL. any treatment, and persistent growth of his liver tumor burden The reaction was allowed to proceed in an Optiplate 384 from was observed. The patient died within a short time period PerkinElmer for 60 minutes at room temperature. Reactions (14 months) after the initial diagnosis due to complications proceeded within the linear reaction time were then terminated of his progressive disease (Fig. 1A). by the addition of 20 mmol/L EDTA and 4 nmol/L Eu-W1024– labeled PY20 antibody. After an incubation of at least 60 minutes, Genomic analyses the samples were excited with a Light Unit laser at 337 nm, and the Discovery of the clonal, KDR/VEGFR2 L840F somatic mutation emission of the LANCE Eu/APC (615/665 nm) was measured by WES-cfDNA. WES-cfDNA confirmed the WT status of the with an Envision reader (PerkinElmer). To test the effect of known KRAS/NRAS/PIK3CA/BRAF genes and uncovered two known VEGFR2 inhibitors on kinase activity, 0.3 ng WT VEGFR2 and 300 colorectal cancer driver mutations, APC c.3964G>T E1322X ng mutant VEGFR2 were used. The starting concentration of the (COSM18702) and TP53 c.659A>C Y220S (COSM43850; Sup- inhibitors tested was 10 mmol/L, followed by 1:5 serial dilutions. plementary Figs. S1 and S2). Interestingly, we identified the novel > In order to calculate IC50 values of inhibition versus DMSO, the KDR c.2518C T mutation leading to the VEGFR2 L840F mutation data were plotted against the inhibitor concentration and fitted to at the protein level (Fig. 1B). The mutation was confirmed in the a sigmoid dose-response curve using the Activity base software patient's basal and on-treatment cfDNA samples (collected after from IDBS. progression with FOLFIRI-cetuximab) by TaqMan genotyping assay, but not in the corresponding gDNA, confirming its somatic Immunohistochemistry status (Fig. 1B; Supplementary Fig. S2). Importantly, the mutated Avatar tumor samples were fixed in 10% neutral buffered allele frequencies (MAF) of the KDR/VEGFR2 mutation in the formalin (4% formaldehyde in solution) and paraffin-embedded. tumor were similar to those of trunk colorectal cancer mutations, Subsequently, 3-mm-thick sections were cut from the samples, such as those in APC and TP53 (approximately 30%, 50%, and mounted in superfrost plus slides, and dried overnight. Before 50%, respectively). The MAFs of KDR/VEGFR2, APC, and TP53 staining, the sections were deparaffinized in xylene and rehy- mutations in plasma were 11%, 8%, and 18%, respectively. drated through a series of decreasing ethanol concentration in water. Consecutive sections were stained with hematoxylin and High concordance between WES-cfDNA and WES-tumor. The con- eosin and by immunohistochemistry, using an automated immu- cordance between WES-cfDNA and WES performed in tDNA nostaining platform (Ventana Discovery XT, Roche, or Autostai- (WES-tumor) obtained after a second liver metastasis was 73% ner Plus Link 48). Antigen retrieval was first performed with high (54/74 somatic mutation, Fig. 2A). Importantly, WES-cfDNA or low pH buffer (CC1m, Roche), endogenous peroxidase was was able to detect a variety of mutation types, such as frameshift blocked (3% hydrogen peroxide), and the slides were incubated (including insertions and deletions), missense, noncoding with an anti p-ERK rabbit polyclonal primary antibody (1:300; (splicing), and nonsense mutations. In addition, WES-cfDNA fi fi Cell Signaling Technology; cat. Nr 9101) for 28 minutes. Subse- discovered 14 high-con dence somatic mutations not identi ed quently, the slides were incubated with the corresponding visu- by WES-tumor. Overall, in the absence of a tumor specimen, WES- alization system (OmniRabbit, Ventana, Roche) with signal cfDNA could identify 68 of the 88 (77.3%) total somatic muta- fi amplification conjugated with horseradish peroxidase. The signal tions identi ed by both techniques. The complete list of the fi was developed using 3,30-diaminobenzidine tetrahydrochloride identi ed somatic mutations, and all the sequencing parameters (DAB) as a chromogen (Chromomap DAB, Ventana, Roche or and genomic annotation are depicted in Supplementary Table S1. DAB solution, Dako), whereas the nuclei were counterstained High concordance between WES-cfDNA and WES-tumor was with Carazzi's hematoxylin. Finally, the slides were dehydrated, also observed in copy-number variation analysis (Fig. 2B). cleared, and mounted with a permanent mounting medium for microscopic evaluation. The entire slide was scanned with a slide KDR/VEGFR2 somatic mutations are recurrent in cancer. KDR scanner (Axio Z1, Zeiss), and images were captured with the ZEN bona-fide cancer mutations (polymorphisms excluded) were software (Zeiss) after evaluation by a trained veterinary pathol- found in all publicly available cancer databases as follows: ogist. Image analysis and quantification were performed using the COSMIC v80 (691/33,320; 1.6%); GENIE (1,012/38,207; AxioVision software package (Zeiss). 2.6%); PCAWG (4/35; 11.5%); and MSK-IMPACT (266/

3554 Clin Cancer Res; 24(15) August 1, 2018 Clinical Cancer Research

Exome Sequencing of Plasma DNA and VEGFR2 Somatic Mutations

A Whole-exome sequencing of plasma cfDNA 80

MAFs 20

0 CIC F11 RP1 DSP TTN APC KDR ND5 IPO9 REST APLF SPTB MN1 SACS TP53 KRT5 ERC1 RBP3 RTN4 CHD7 APOB FSTL1 RBPJL TMX2 MTA1 MARS TESK1 MDC1 PIANP FLAD1 CECR2 NLRP9 FOXP1 SCRN1 NBPF1 NBPF1 NBPF1 PRIM2 CDC45 OR5R1 EIF2B5 AGFG2 PPFIA1 OR8G1 OR8G1 NSUN7 MYO7B OR2M4 OR5M3 WDR89 WDR89 WDR89 WDR89 WDR89 ZNF492 ATP8B3 TRIM43 TBC1D9 NBPF10 PABPC3 PABPC3 PABPC3 PABPC3 PABPC3 PABPC3 C2orf81 HECTD4 BOD1L2 BOD1L2 CHAF1A NECAB2 PCDH15 PDGFRA MRPS31 MAP3K5 DYNC1I1 MCOLN2 NPBWR1 RAD21L1 PPP1R3A COL26A1 OLFML2A HDGFRP2 KIAA1429 FAM120B TMEM159 TMEM198 C14orf169 WES plasma cfDNA WES tumor DNA

B Tumor metastasis

Pretreatment plasma cfDNA

C D EGFR R841K/R; KIT R796A/G/P/K; ALK R1253G/T; AXL R676M; CSK R318C; EPHA2 BRAF p.V600E/K/D/R/M/L/A/Q/V (44,637) R743G/H/R; EPHA3 R750L/Q/W; EPHA7 FLT3 p.D835E/Y/F/E/H/V/N/A (1,367) % of KDR somac EGFR p.S768I/T/C/G/N (214) R762C; EPHB1 R748K/S; EPHB2 R750C; JAK3 KIT p.D816V/E/Y/H/F/A/I/N/G (2,019) R651Q; MERTK R727Q; NTRK3 R683S; EGFR p.L861Q/R/K/E/F/P/V/L (405) mutaons PDGFRA R822F/H; ROR1 R619C/H/S PDGFRA p.D842H/E/V/I/Y/A (544) 3.5 3 BRAF p.G466V/E/R/A (70) EGFR p.L858R/A/G/K/M/P/Q/V/W (9,899) 2.5 EGFR p.G721A/D/S/V/W (7) BRAF p.L597R/Q/S/L/P/V (90) 2 1.5 EGFR p.L718L/V/P/M (9) BRAF p.K601E/T/K/N/Q/I/L (193) 1 G843D I892T L1049W D1052N/G/H 0.5 ABL1 p.L248R/V (23) 0 R1032Q I1053M S1100F L840F S925F R1022Q

aa438 VEGFR2 tyrosine kinase domain 0611 aa

Figure 2. Concordance between the genomic landscape identified by whole-exome sequencing of plasma cfDNA and tumor; DNA and recurrence of KDR/VEGFR2 oncogenic mutations in human cancers. A, The histograms represent all the genes with somatic mutations identified by plasma cfDNA whole-exome sequencing (in red) and by tumor whole-exome sequencing (in blue). The list and genomic annotation of all mutations are shown in SupplementaryTable S1. The mutated allele frequencies (MAFs) of each mutation in relation to all reads are depicted in Y. B, Copy number alteration landscape portrayed by the whole-exome sequencing of tumor DNA (upper part) and pre-treatment plasma cfDNA (lower part). Gains are depicted in green, losses in red, and normal (balanced) in grey. C, Frequency of KDR somatic mutations in large cancer genomic sequencing projects including more than 70,000 cancer samples. Common germline polymorphisms were excluded from these analyses and only cancer-exclusive mutations were considered. D, Representative results from the search of genomics and protein databases, showing several mutations, structurally analogous to those of KDR, identified in other cancer-relevant kinases.

10,336; 2.6%; Fig. 2C). Importantly, 15 of the 35 (43%) colorectal P/K, ALK R1253G/T, AXL R676M, CSK R318C, EPHA2 R743G/H/ cancer samples in the PCAWGS study had KDR gene ampliﬁca- R, EPHA3 R750L/Q/W, EPHA7 R762C, EPHB1 R748K/S, EPHB2 tion. KDR somatic mutations found in the cancer genomics R750C, JAK3 R651Q, MERTK R727Q, NTRK3 R683S, PDGFRA databases are shown in Supplementary Table S5. R822F/H, and ROR1 R619C/H/S (Fig. 2D).

KDR/VEGFR2 somatic mutations are analogous to known cancer- Functional studies of the VEGFR2 cancer mutants related kinase mutations. KDR/VEGFR2 cancer mutations occur in VEGFR2 L840F causes cancer therapy resistance. In agreement with hot-spot residues analogous to those of other kinases related to the phenotype we observed in our patient, the patient-derived human cancers. For example, VEGFR2 L1049W is analogous to Avatar model (AvatarVEGFR2:L840F) did not respond to multiple EGFR L858, and VEGFR2 D1052N/G/H is analogous to FLT3 anti-VEGF and VEGFR2 inhibitors, whereas the AvatarVEGFR2:WT D835, KIT D816, EGFR L861, and PDGFRA D842. Importantly, model, used as a control, was sensitive to all the treatments we found that R1032Q (COSM192176) is the most frequent (Fig. 3A). Our 3D structural analysis revealed that L840 is VEGFR2 mutation in the cancer databases and possibly a muta- located exactly at the entrance of the ATP-binding pocket of tional hot-spot of VEGFR2 because it is analogous to a number of the tyrosine kinase domain of VEGFR2 (aa840-LGXGXXG- known cancer mutations, such as EGFR R841K/R, KIT R796A/G/ 846aa), and it forms hydrophobic interactions with many

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3555

Toledo et al.

A PDX model tumor inhibion 80 70 WT L840F 60 50 40 30 20 vs . untreated 10

% Growth inhibion 0

* * *

B ATP-binding pocket VEGFR2 Resistant-causing analogous mutaons VEGFR2 L840F ABL1 L248V EGFR L718Q/R

Axitinib Cabozantinib Dovitinib Lenvatinib

Log (µmol/L) Log (µmol/L) Log (µmol/L) Log (µmol/L)

Figure 3. The VEGFR2 L840F mutant leads to broad and strong cancer therapy resistance and R1032Q mutant to increased sensitivity to VEGFR2 inhibitors. A, Growth inhibition of the patient-derived xenograft (PDX) Avatar model carrying the KDR/VEGFR2 L840F (red) after 3 weeks of treatment with anti-VEGF drugs (B20/murine and bevacizumab/human), VEGFR2 kinase inhibitors (axitinib, cabozantinib, cabozantinib:MEK inhibitor combo, pazopanib, regorafenib, sorafenib), or inhibitors of other kinases, such as afatinib (EGFR), crizotinib (MET), and MEK inhibitor (MEKi). A second CRC PDX model, carrying the KDR/VEGFR2 WT (blue), was used as a positive control, and was treated with some of the above drugs. To analyze the inhibition of tumor growth promoted by each tested drug, tumor volumes of the untreated mice were set as 100% growth and used as reference for the measurement of the treated animals (tumor volume of treated mice divided by tumor volume of untreated mice). B, Localization of structurally analogous L residue mutations in VEGFR2 (L840), EGFR (L718), and ABL1 (L248). A close-up view of the entrance of the ATP-binding pocket domain is shown in green for VEGFR2 L840F, in light blue for ABL1 L248V, and in orange for EGFR L718Q/R. Patients with these mutations are all refractory to treatment with tyrosine kinase inhibitors, which directly bind to these L residues (see also Supplementary Fig. S3). C, Inhibition of kinase activity of WT (black), L840F (red), and R1032Q (blue) VEGFR2 kinase domains by VEGFR2 inhibitors axitinib, cabozantinib, dovitinib, and lenvatinib. WT VEGFR2 kinase was sensitive to the four inhibitors, especially to axitinib and cabozantinib. The concentration of the mutants was increased 1,000 times to achieve a measurable kinase activity, which was consequently measured in the presence of TKIs.

FDA-approved small-molecule kinase inhibitors (Fig. 3B; Sup- model, in vitro kinase assays with recombinant VEGFR2 kinase plementary Fig. S3). Molecular dynamics simulations of the domains showed that WT VEGFR2 has high kinase activity, L840F-mutant VEGFR2 also show that most of the conforma- whereas L840F VEGFR2 has impaired kinase activity, suggesting tions of F840 observed in the simulations are not compatible possible loss of ATP binding (Supplementary Fig. S4A–S4B). with inhibitor binding. Consistent with the computational Consistent with these results, Y1175 phosphorylation of L840F

3556 Clin Cancer Res; 24(15) August 1, 2018 Clinical Cancer Research

Exome Sequencing of Plasma DNA and VEGFR2 Somatic Mutations

Figure 4. Oncogenic potential of VEGFR2 cancer mutants in xenograft assays. Colo320 CRC cell lines were used for xenograft studies, as they resemble the mutation proﬁling of the studied patient’s tumor, with wild-type KRAS/NRAS/BRAF/PIK3CA status, and TP53 and APC mutations. Colo320 cells stably expressing different VEGFR2 mutants were injected subcutaneously in four immune deﬁcient mice (each blue bar represents one mouse). Results of xenograft growth after 2 months following injections or when tumors reached the established humane endpoint are shown.

VEGFR2 was significantly reduced compared with WT VEGFR2, or the kinase inactive dominant-negative K868M VEGFR2 did not both in human embryonic kidney (HEK293) cells transiently generate tumors within 120 days after cell injections. Our data transfected with WT or L840F VEGFR2 and in porcine endo- suggest that cancer-associated VEGFR2 mutants might have onco- thelial (PAE) cells stably expressing WT or L840F VEGFR2 genic potential. (Supplementary Fig. S4C–S4D). Our kinase activity assays showed that whereas WT VEGFR2 was sensitive to axitinib, Discussion cabozantinib, dovitinib, and lenvatinib, L840F VEGFR2 was resistanttoallthesedrugs,albeitatdistinctlevels(Fig.3C). The high capacity of WES-cfDNA for portraying the somatic mutation and copy-number variation landscapes of tumors VEGFR2 R1032Q confers sensitivity to strong VEGFR2 inhibitors. (77.3% of concordance rate in our case) shows an immense Kinase assay showed that similar to the L840F, the R1032Q potential for research in translational oncology (12–14). We mutation greatly reduced VEGFR2 kinase activity (Supplementary anticipate that the implementation of WES-cfDNA will enable Fig. S4B), although the molecular mechanism underlying this blood-based global genomic profiling of cancers and greatly phenotype would be distinct. As R1032Q affects the universal potentiate the discovery of new biomarkers. Moreover, it has the kinase catalytic motif DxxxxN (aa1028-DxxxRN-1033aa; ref. 11), potential to be used in clonal evolution and tumor mutation and not the ATP-binding site as the L840F, we investigated burden analyses. whether R1032Q VEGFR2 would be inhibited by TKIs. In vitro Applying WES-cfDNA, we identified the VEGFR2 L840F clonal, kinase assays showed increased sensitivity of R1032Q VEGFR2 somatic mutation in a highly refractory mCRC patient. Our to TKIs (Fig. 3C; Supplementary Fig. S5A). Furthermore, prolif- validation experiments with cell lines, animal models, and bio- eration studies with the Colo-320 colorectal cell line, which chemical assays demonstrated that VEGFR2 mutants can modu- has a similar mutation profile as that of the patient's tumor late the response to antiangiogenic agents according to their (WT KRAS/NRAS/BRAF/PIK3CA status, and TP53 and APC muta- localization and functional consequence. For example, the tions), showed that stable expression of R1032Q VEGFR2 con- L840F ATP-binding pocket domain mutation causes very strong ferred sensitivity to lenvatinib [growth inhibition (GI50) ¼ 20.8 and broad resistance to anti-VEGF and VEGFR2 inhibitors, where- for the R1032Q compared with 36.4 for WT VEGFR2] and to as the R1032Q kinase domain mutation is apparently sensitive to cabozantinib (GI50 ¼ 2.5 for the R1032Q compared with 7.9 for strong VEGFR2 inhibitors, such as cabozantinib and lenvatinib. WT VEGFR2; Supplementary Fig. S5B). Moreover, cabozantinib These findings demonstrate for the first time that the same well- treatment of the MDST8 colorectal cancer cell line, naturally known genetic mechanisms leading to resistance/increase sensi- harboring the KDR R1032Q mutation, led to a prominent tivity to inhibitors of EGFR, ABL1, and PDGFRA receptors do also decrease in cell growth rate in vitro and diminished the high occur in VEGFR2 and can have implications for the outcome of constitutive ERK phosphorylation levels. Importantly, we found antiangiogenic treatments (15). that such downstream inhibition was specific to cabozantinib, a In agreement with our preclinical experimental findings, very strong VEGFR2 (0.035 nmol/L) and c-MET (1.3 nmol/L) Knepper and colleagues very recently reported a prolonged inhibitor, and occurred in cells treated in the absence or presence complete response to pazopanib in a metastatic basal cellular of VEGF (Supplementary Fig. S5A). carcinoma patient carrying the KDR/VEGFR2 R1032Q somatic mutation (16). Interestingly, a second mCRC patient case with VEGFR2 cancer–related mutants are oncogenic. We further asked if the KDR/VEGFR2 R961W somatic mutation was also recently in addition to modulating the response to TKIs, VEGFR2 cancer reported. After progression to 5-fluorouracil-bevacizumab, the mutants could promote tumor growth. Indeed, we showed that patient responded to low-dose regorafenib with remarkable when Colo-320 cells stably expressing D717V, G800D, G800R, regression of the hepatic metastases, abdominal and retroper- L840F, G843D, S925F, R1022Q, R1032Q, and S1100F VEGFR2, itoneal lymph nodes, and rectosigmoid colon hypermetabolic when injected in mice, even in small numbers and without lesions (17). Matrigel, could generate tumors that reached the established Another key point of our study relates to the emerging role humane endpoint (Fig. 4). Importantly, Colo-320–expressing EV of loss-of-function kinase mutations to tumorigenesis and

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3557

Toledo et al.

cancer therapy modulation. We investigated extensively the Authors' Contributions activity of L840F and R1032Q VEGFR2, and the results from Conception and design: R.A. Toledo, E. Garralda, S. Perea, C. Blanco-Aparicio, different and independent experiments clearly show that both A. Cubillo, J.L. Martínez-Torrecuadrada, M. Hidalgo mutants are kinase-dead. These findings are in line with Development of methodology: R.A. Toledo, J. Monsech, A. Otero, N. Banos,~ previous work from Dr. Owen Samson's and Dr. John Y. Duran, F. Sarno, S. Perea, A. De Martino, A. Cubillo, O. Domínguez, M. Hidalgo Brognard's groups showing that recurrent MLK4 and PKC Acquisition of data (provided animals, acquired and managed patients, mutations in colorectal cancer are both loss of function and provided facilities, etc.): R.A. Toledo, E. Garralda, M. Mitsi, J. Monsech, oncogenic (18, 19). Importantly, recent studies demonstrated E. Vega, A. Otero, M.I. Albarran, N. Banos,~ F. Sarno, T. Sanchez-Perez, S. Perea, that BRAF-impaired cancer mutants could recruit CRAF, pro- R. Alvarez, A. Cubillo, O. Domínguez, J.L. Martínez-Torrecuadrada, M. Hidalgo moting alternative MAPK pathway activation and leading to Analysis and interpretation of data (e.g., statistical analysis, biostatistics, BRAF inhibitor therapy resistance. Interestingly, this BRAF: computational analysis): R.A. Toledo, E. Garralda, M. Mitsi, T. Pons, CRAF pathway rewiring confers de novo clinical sensitivity to M. Camacho-Artacho, S. Perea, A. De Martino, D. Lietha, C. Blanco-Aparicio, dasatinib (ABL, SRC, and c-Kit inhibitor; ref. 20). The VEGFR2 M. Hidalgo Writing, review, and/or revision of the manuscript: R.A. Toledo, E. Garralda, L840F and R1032Q kinase-impairing mutations would fit M. Mitsi, T. Pons, S. Perea, R. Alvarez, A. De Martino, D. Lietha, well in such a "rewiring mutant kinase oncogenic model C. Blanco-Aparicio, A. Cubillo, J.L. Martínez-Torrecuadrada, M. Hidalgo (Supplementary Fig. S6)." However, tumor growth of the Administrative, technical, or material support (i.e., reporting or organizing VEGFR2:L840F Avatar model was not significantly reduced upon data, constructing databases): R.A. Toledo, E. Garralda, V. Bonilla, S. Perea treatment with inhibitors of EGFR (afatinib), MET (crizotinib), Study supervision: R.A. Toledo, E. Garralda, A. Cubillo, M. Hidalgo andMAPK(MEKi;Fig.3A)andthepossiblerewiringpartners of VEGFR2 L840F remain unknown. Acknowledgments In summary, the current study highlights the capability of R.A. Toledo holds a Miguel Servet-I research contract by Institute of Health "Carlos III" of the Ministry of Economy (CP17/00199) and exomic sequencing of cfDNA from plasma of cancer patients as Competitiveness and is supported by a Fundacíon Olga Torres a powerful platform for portraying the somatic mutation land- emerging researcher grant. The authors are grateful to Dr. Pedro scape of cancer and discovering new mechanisms of resistance to P. Lopez-Casas and Manuel Munoz~ (CNIO Gastrointestinal Cancer cancer therapies. Because of its advantage to generate results Unit) for their valuable technical and administrative assistance and highly concordant to those of tumor sequencing without the to Kurt Ballmer-Hofer (Paul Scherrer Institute, Switzerland) for his hurdle of conventional tumor biopsies, we anticipate that involvement at the beginning of the functional in vitro experiments. We would like to thank especially the patient and his family for their WES-cfDNA will become frequently used in oncology. Moreover, participationinthestudy. our study characterized KDR/VEGFR2 somatic mutations as potential genetic biomarkers of response to antiangiogenic cancer The costs of publication of this article were defrayed in part by the payment of therapies, and these findings may serve as reference for further page charges. This article must therefore be hereby marked advertisement in studies on the topic. accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Disclosure of Potential Conflicts of Interest Received January 10, 2018; revised February 15, 2018; accepted March 21, No potential conflicts of interest were disclosed. 2018; published first March 29, 2018.

References 1. Arnold M, Soerjomataram I, Ferlay J, Forman D. Global incidence of 9. Toledo RA, Qin Y, Cheng ZM, Gao Q, Iwata S, Silva GM, et al. Recurrent oesophageal cancer by histological subtype in 2012. Gut 2015;64:381–7. mutations of chromatin-remodeling genes and kinase receptors in pheo- 2. Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim chromocytomas and paragangliomas. Clin Cancer Res 2016;22:2301–10. W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for 10. Garralda E, Paz K, Lopez-Casas PP, Jones S, Katz A, Kann LM, et al. metastatic colorectal cancer. N Engl J Med 2004;350:2335–42. Integrated next-generation sequencing and avatar mouse models for per- 3. Wagner AD, Arnold D, Grothey AA, Haerting J, Unverzagt S. Anti- sonalized cancer treatment. Clin Cancer Res 2014;20:2476–84. angiogenic therapies for metastatic colorectal cancer. Cochrane Database 11. Hanks SK, Hunter T. The eukaryotic protein kinase superfamily: kinase Syst Rev 2009;1:1–75. (catalytic) domain structure and classification. FASEB J 1995;9:576–96. 4. Matos I, Elez E, Capdevila J, Tabernero J. Emerging tyrosine kinase inhi- 12. Murtaza M, Dawson SJ, Tsui DW, Gale D, Forshew T, Piskorz AM, et al. bitors for the treatment of metastatic colorectal cancer. Expert Opin Emerg Non-invasive analysis of acquired resistance to cancer therapy by sequenc- Drugs 2016;21:267–82. ing of plasma DNA. Nature 2013;497:108–12. 5. Lambrechts D, Claes B, Delmar P, Reumers J, Mazzone M, Yesilyurt BT, et al. 13. Butler TM, Johnson-Camacho K, Peto M, Wang NJ, Macey TA, Korkola JE, VEGF pathway genetic variants as biomarkers of treatment outcome with et al. Exome sequencing of cell-free DNA from metastatic cancer patients bevacizumab: an analysis of data from the AViTA and AVOREN rando- identifies clinically actionable mutations distinct from primary disease. mised trials. Lancet Oncol 2012;13:724–33. PLoS One 2015;10:e0136407. 6. Loupakis F, Cremolini C, Yang D, Salvatore L, Zhang W, Wakatsuki T, et al. 14. Adalsteinsson VA, Ha G, Freeman SS, Choudhury AD, Stover DG, Parsons Prospective validation of candidate SNPs of VEGF/VEGFR pathway in HA, et al. Scalable whole-exome sequencing of cell-free DNA reveals high metastatic colorectal cancer patients treated with first-line FOLFIRI plus concordance with metastatic tumors. Nat Commun 2017;8:1324. bevacizumab. PLoS One 2013;8:e66774. 15. Ercan D, Choi HG, Yun CH, Capelletti M, Xie T, Eck MJ, et al. EGFR 7. Chatterjee S, Heukamp LC, Siobal M, Schottle€ J, Wieczorek C, Peifer M, et al. mutations and resistance to irreversible pyrimidine-based EGFR inhibitors. Tumor VEGF:VEGFR2 autocrine feed-forward loop triggers angiogenesis in Clin Cancer Res 2015;21:3913–23. lung cancer. J Clin Invest 2013;123:1732–40. 16. Knepper TC, Freeman ML, Gibney GT, McLeod HL, Russell JS. Clinical 8. Toledo RA, Cubillo A, Vega E, Garralda E, Alvarez R, de la Varga LU, et al. response to pazopanib in a patient with KDR-mutated metastatic basal cell Clinical validation of prospective liquid biopsy monitoring in patients carcinoma. JAMA Dermatol 2017;153:607–9. with wild-type RAS metastatic colorectal cancer treated with FOLFIRI- 17. Loaiza-Bonilla A, Jensen CE, Shroff S, Furth E, Bonilla-Reyes PA, Deik AF, cetuximab. Oncotarget 2017;8:35289–300. et al. KDR mutation as a novel predictive biomarker of exceptional

3558 Clin Cancer Res; 24(15) August 1, 2018 Clinical Cancer Research

Exome Sequencing of Plasma DNA and VEGFR2 Somatic Mutations

response to regorafenib in metastatic colorectal cancer. Cureus 2016;8: 19. Antal CE, Hudson AM, Kang E, Zanca C, Wirth C, Stephenson NL, et al. e478. Cancer-associated protein kinase C mutations reveal kinase's role as tumor 18. Marusiak AA, Stephenson NL, Baik H, Trotter EW, Li Y, Blyth K, suppressor. Cell 2015;160:489–502. et al. Recurrent MLK4 loss-of-function mutations suppress JNK 20. Sen B, Peng S, Tang X, Erickson HS, Galindo H, Mazumdar T, et al. Kinase- signaling to promote colon tumorigenesis. Cancer Res 2016;76: impaired BRAF mutations in lung cancer confer sensitivity to dasatinib. 724–35. Sci Transl Med 2012;4:136ra70.

www.aacrjournals.org Clin Cancer Res; 24(15) August 1, 2018 3559

Exome Sequencing of Plasma DNA Portrays the Mutation Landscape of Colorectal Cancer and Discovers Mutated VEGFR2 Receptors as Modulators of Antiangiogenic Therapies

Rodrigo A. Toledo, Elena Garralda, Maria Mitsi, et al.

Clin Cancer Res 2018;24:3550-3559. Published OnlineFirst March 27, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-18-0103

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2018/06/05/1078-0432.CCR-18-0103.DC2

Cited articles This article cites 20 articles, 6 of which you can access for free at: http://clincancerres.aacrjournals.org/content/24/15/3550.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/24/15/3550.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://clincancerres.aacrjournals.org/content/24/15/3550. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.