Identiﬁcationof KIF5B-RET and GOPC-ROS1 FusionsinLung ... · the nononcogenic, reciprocal fusion...

Human Cancer Biology

IdentificationofKIF5B-RETandGOPC-ROS1Fusions in LungAdenocarcinomas through a Comprehensive mRNA-BasedScreen for Tyrosine Kinase Fusions

Yoshiyuki Suehara1, Maria Arcila1, Lu Wang1, Adnan Hasanovic1, Daphne Ang1, Tatsuo Ito1, Yuki Kimura1,Alexander Drilon2, Udayan Guha6, Valerie Rusch3, Mark G. Kris2, Maureen F. Zakowski1, Naiyer Rizvi2,Raya Khanin4, and Marc Ladanyi1,5

AbstractBackground: The mutually exclusive pattern of the major driver oncogenes in lung cancer suggests that

other mutually exclusive oncogenes exist. We conducted a systematic search for tyrosine kinase fusions by

screening all tyrosine kinases for aberrantly high RNA expression levels of the 30 kinase domain (KD) exons

relative to more 50 exons.Methods: We studied 69 patients (including five never smokers and 64 current or former smokers) with

lungadenocarcinomanegative for allmajormutations inKRAS,EGFR,BRAF,MEK1,HER2, and forALK fusions

(termed "pan-negative"). ANanoString-based assaywas designed to query the transcripts of 90 tyrosine kinases

at two points: 50 to the KD and within the KD or 30 to it. Tumor RNAs were hybridized to the NanoString

probes and analyzed for outlier 30 to 50 expression ratios. Presumed novel fusion events were studied by

rapid amplification of cDNA ends (RACE) and confirmatory reverse transcriptase PCR (RT-PCR) and FISH.

Results:We identified one case each of aberrant 30 to 50 ratios in ROS1 and RET. RACE isolated a GOPC-

ROS1 (FIG-ROS1) fusion in the former and aKIF5B-RET fusion in the latter, both confirmedbyRT-PCR. The

RET rearrangement was also confirmed by FISH. The KIF5B-RET patient was one of only five never smokers

in this cohort.

Conclusion: The KIF5B-RET fusion defines an additional subset of lung cancer with a potentially

targetable driver oncogene enriched in never smokers with "pan-negative" lung adenocarcinomas. We also

report in lung cancer the GOPC-ROS1 fusion originally discovered and characterized in a glioma cell line.

Clin Cancer Res; 18(24); 6599–608. �2012 AACR.

IntroductionThe management of lung adenocarcinoma has been

transformed by the discovery of targetable driver oncogenessuch as EGFRmutations and ALK fusions (1, 2). It has alsobecome apparent that these major driver oncogenes, alongwith KRAS, HER2, and others, are mutually exclusive.Indeed, this remarkable, mutually exclusive pattern ofmutations suggests that other mutually exclusive driveroncogenes remain to be discovered. On the basis of exten-sive genotyping data generated by our group and others

(3–5), it is estimated that approximately 60% of lungadenocarcinomas in the United States contain one of thefollowing known driver oncogenes, all mutually exclusivewith rare exceptions:KRAS, EGFR,ALK fusion,BRAF,HER2,NRAS, and MEK1. This pattern of mutations in more than60% of cases suggests a general dependence of lung ade-nocarcinoma on mutational activation of certain down-stream signaling pathways [mitogen-activated proteinkinase (MAPK), AKT] and provides a rationale for a system-atic screen of the remaining, "pan-negative" cases for otherdriver oncogenes. As part of such a screening effort, we haveused an RNA-based approach to systematically search forevidence of novel tyrosine kinase fusions in "pan-negative"lung adenocarcinomas.

The overall rationale for our exon-level mRNA-basedscreen for the detection of fusion genes is based on theobservation that most gene fusions that lead to theformation of a chimeric fusion protein cause an intra-genic discontinuity in the RNA expression level of theexons that are 50 or 30 to the fusion point in one or both ofthe fusion partners. This is attributable to the differencesin the strength or activity of the promoters of the 2translocation partner genes. In addition, in some cases,

Authors' Affiliations: Departments of 1Pathology, 2Medicine, and 3Sur-gery, 4Bioinformatics Core, 5HumanOncology andPathogenesis Program,Memorial Sloan-Kettering Cancer Center, New York, New York; and6National Cancer Institute, Bethesda, Maryland

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Corresponding Author: Marc Ladanyi, Department of Pathology,Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, NewYork, NY 10065. Phone: 212-639-6369; Fax: 212-717-3515; E-mail:[email protected]

doi: 10.1158/1078-0432.CCR-12-0838

�2012 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 6599

on March 11, 2021. © 2012 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 10, 2012; DOI: 10.1158/1078-0432.CCR-12-0838

http://clincancerres.aacrjournals.org/

the nononcogenic, reciprocal fusion gene may be lost dueto an unbalanced translocation event. We have recentlyshown the successful application of this general strategyin the detection of gene fusions without prior knowledgeof the genetic background of a given case, by identifying anovel, highly recurrent HEY1-NCOA2 fusion in the mes-enchymal subtype of chondrosarcoma based on analysisof Affymetrix Exon Array data (6). Here, we developed anefficient NanoString-based strategy that follows the sameprinciple but is focused on tyrosine kinases as moreimmediately actionable gene fusions. We describe belowhow the application of this comprehensive NanoString-based assay for tyrosine kinases with outlier 50 to 30

expression ratios in "pan-negative" lung adenocarcino-mas led to the identification of novel KIF5B-RET andGOPC-ROS1 fusions.

Materials and MethodsAssay validation samples and negative control samples

To validate the performance of the NanoString assaydesign, we used cell lines and patient tumor samples withknown fusions. The cell lines included H2228 and H3122(both EML4-ALKþ lung adenocarcinomas), K299 (NPM1-ALKþ anaplastic large cell lymphoma), HCC78 (CD74-ROS1þ lung adenocarcinoma), and U118 (FIG-ROS1þglioma). The patient tumor samples, tested underMemorialSloan-Kettering Cancer Center (MSKCC; New York, NY)Institutional Review Board (IRB) protocols as described inmore detail below, included 75 lung adenocarcinoma sam-ples of which 24were positive for, and 51were negative for,evidence of ALK fusion by ALK FISH and at least one othermethod, either EML4-ALK real-time PCR (RT-PCR) orimmunohistochemistry (IHC) using the D5F3 ALK mono-clonal antibody (Cell Signaling). As negative control sam-ples to establish the range of normal 50:30 expression ratio

variability for each gene in the NanoString assay, we includ-ed 17 KRAS-mutated lung adenocarcinomas, 11 EGFR-mutated lung adenocarcinomas, and 37 samples of non-neoplastic lung tissue procured at time of lung cancersurgery.

Discovery samplesWe identified 69 "pan-negative" lung adenocarcinomas

with frozen tumor available for RNA extraction based ondata from ongoing reflex clinical genotyping of resectedlung adenocarcinomas at MSKCC (4). These samples werenot selected for demographic features or smoking history,and these data are listed in Supplementary Table S1. Thesummary of the data for these parameters are as follows: 64smokers (50 former smokers including 5 with 5 pack-yearsor less, 14 current) and 5 never smokers (all female); 66white non-Hispanic; and3Hispanic or Black. Therewere noAsian patients in this study population. The genotypingprocess used to identify these samples as "pan-negative"for all known lung adenocarcinoma driver oncogenes isoutlined in Fig. 1 and briefly summarized here: tumorsampleswere subjected to extendedmolecular testing underMSKCC IRB protocols, consisting of a prospective screenfor all recurrent mutations in 8 key genes (EGFR, HER2,KRAS, NRAS, BRAF, MEK1, PIK3CA, AKT1) in all lungadenocarcinomas using the Sequenom platform. The panel

• Reflex testing of lung adenocarcinomas resected at MSKCC

and banking of frozen tumor and matched normal tissue

If (-)

• EGFR mutation testing for exon 19 insertions, and L858R

mutation

If(-)

• KRAS mutation testing for point mutations in exon 2 (codons

12-13)

If (-)

• ALK fusion testing by FISH

If (-)

• Sequenom assays for major point mutations (total: 91) in EGFR, KRAS, NRAS, BRAF, HER2, PIK3CA, MEK1, and AKT

If (-)

• Multiplex sizing assay for insertions in EGFR Exon 20 and

HER2 exon 20

If (+)

• NanoString assay for tyrosine kinases with outlier 5’:3’

expression ratios

• 5’ RACE analysis of fusion candidates

Figure 1. Overall strategy for identification of "pan-negative" tumors fordiscovery of novel tyrosine kinase fusions. Note that samples withPIK3CA mutations were not excluded from further analysis because oftheir known frequent overlap with other major driver mutations. SeeMaterials and Methods for further details.

Translational RelevanceThe identification of druggable driver oncogenes such

asmutated EGFR andHER2 andALK fusions represents amajor advance in the therapy of lung adenocarcinoma.However, the majority of patients with lung adenocar-cinoma still do not have a targeted treatment option.Here, we focused on patients with lung adenocarcinomawithout any identifiable driver oncogenes to search fornew potential targets. In this enriched patient popula-tion, we used a NanoString-based strategy to search fornew tyrosine kinase fusion genes.With this approach,weidentified and confirmed one case each of a novelKIF5B-RET fusion and a GOPC-ROS1 (FIG-ROS1). AlthoughRET and ROS1 fusions represent small percentages oflung adenocarcinoma, they are of immediate clinicalinterest. ROS1 fusions have been shown to respondclinically to targeted treatment with crizotinib, andpreclinical data suggest that RET fusions shouldrespond to RET inhibitors.

Suehara et al.

Clin Cancer Res; 18(24) December 15, 2012 Clinical Cancer Research6600




interrogates these 8 genes for the presence of 91 pointmutations (5). EGFR exon 19 deletions, exon 20 insertions,andHER2 exon 20 insertionswere screened by PCRproductsizing assays (7, 8). Cases containing PIK3CA mutations,even as the sole detectable mutation, were not excludedbecause PIK3CA mutations are known to frequently co-occur with other driver mutations (9). Cases negative withthe preceding assays (except PIK3CA) were then screened byFISH for evidence of ALK fusions using the Abbott/VysisALK breakapart FISH assay.

NanoString assay for kinase fusionsThe NanoString nCounter system is a fluorescence-

based platform to detect individual mRNA moleculeswithout PCR amplification in a quantitative and highlymultiplexed fashion (10, 11). In this system, each captureprobe and reporter probe together query a contiguous100-bp region and only 100 ng RNA is needed perreaction. Our NanoString assay design (Fig. 2) was basedupon the known genomic properties of existing tyrosinekinase fusions, namely, that these fusions typically occurupstream of the exons encoding the kinase domain. Theexons encoding the kinase domain GXGXXG motif for all90 tyrosine kinases and 3 serine/threonine kinases (BRAF,ARAF, CRAF) were identified as described elsewhere (12).All exons were labeled according to ENSEMBL number-ing. On the basis of this mapping, 2 100-bp regions wereselected for each gene transcript, a 50 probe pair locatedfar upstream of the kinase domain exons, and the secondlocated within those exons or further 30. The 100-bpregions were selected to straddle exon boundaries to

reduce the risk of interfering signal from gDNA. Detailedsequence information for all the kinase gene targetregions is provided in Supplementary Table S2. Eachsample was analyzed using 100 to 200 ng of total RNAper assay. All kinase genes and control genes were assayedsimultaneously in multiplexed reactions. The raw datawere normalized to the nCounter system spike-in positiveand negative controls in each sample. The normalizedresults are expressed as the relative mRNA level. Nano-String count data were converted to the log2 scale (1 wasadded to all data to avoid problems with zeros). The datawere then normalized with the quantile method using thenormalizeBetweenArrays procedure from the Bioconduc-tor R package. The 30/50 ratios were calculated for eachkinase gene in the assay, and samples with outlier ratioswere visualized on log scale plots. Samples with specifictyrosine kinase genes with 30/50 ratios below �4 on thelog2 scale were considered outliers if only seen in rarepan-negative samples and none of the other controlsamples.

50RACERapid amplification of cDNA ends (RACE) was con-

ducted with the use of 50 RACE system (50 system for RapidAmplification of cDNA Ends, version 2.0, Invitrogen). Theprimers used to identify aberrant ROS1 transcripts in RNAin 50 RACE reaction were ROS1-GSP1 primer (50-GAGGAG-ACCTTCTTACTTATTT-30) for cDNA synthesis and ROS1–GSP2 (50-AAGACAAAGAGTTGGCTGAGCTGCG-30) andROS1-GSP3 (50-CTGGCATAGAAGATTAAAGAATC-30) forthe nested PCR reaction. The primers used to identify

Figure 2. Principle of NanoStringassay for evidence of tyrosine kinasefusions. As functional tyrosinekinase fusions typically occurupstream of the exons encoding thekinase domain, probes weredesigned to measure the expressionat 2 regions for each gene transcript,a 50 probe pair located far upstreamof the kinase domain exons and thesecond located within those exonsor further 30. Kinase fusions oftencause an imbalance in the RNAexpression level of these 2 regionsattributable to stronger activation ofthe promoter of the fusion partnergene and/or, in some cases, loss ofthe nononcogenic, reciprocal fusiongene due to an unbalancedtranslocation event. The 30:50

expression ratios are calculated foreach kinase gene in the assay, andsamples with outlier ratios werevisualized on log scale plots.

Gene B (Tyrosine kinase )

5’ 3’

Ex5 Ex4 Ex3 Ex1 Ex2 Ex10 Ex7 Ex8 Ex6 EEEx9

5’ probe

3’ probe

Tyrosine kinase domain

5’ 3’

Ex5 Ex4 Ex3 Ex1 Ex2 Ex10 Ex7 Ex8 Ex6 Ex9

Tyrosine kinase domain

Gene A (Fusion partner )

5’ 3’

Ex5 Ex4 Ex3 Ex1 Ex2 Ex10 Ex7 Ex8 Ex6 EEx9

5’ 3’

Ex5 Ex4 Ex3 Ex1 Ex2 Ex10 Ex7 Ex8 Ex6 EEx9

3’ probe

p

Break point

Dimerization domain

EE

k k k k pp

3

Ge

in

5’ probe

Gene B (TK)

expression data

Nan

oS

trin

g A

ssa

y

Ex

pre

ss

ion

Le

ve

l

5 5

Dimerization domain

Break point

Novel Kinase Fusions in Lung Cancer

www.aacrjournals.org Clin Cancer Res; 18(24) December 15, 2012 6601




aberrant RET transcripts in RNA in 50 RACE reaction wereRET-GSP1 primer (50-GATGAACGAGCTTCATCTCGGC-30)for cDNA synthesis and RET-GSP 2 (50-GACGTTGAACTCT-GACAGCAGGTCTC-30) and RET-GSP3 (50-CTAGAGTTT-TTCCAAGAACCAAGTTCTTC-30) for the nested PCR reac-tion. For sequencing, the RACE PCR product was clonedusing TOPO TA Cloning Kit (Invitrogen), purified withQiaprep Spin Miniprep Kit (QIAGEN) and sequenced bySanger sequencing.

RT-PCR for KIF5B-RETTo detect the presence of the KIF5B-RET fusion tran-

script, we conducted RT-PCR using several independentsets of forward and reverse primers, to ensure that the RT-PCR products were short (<350 bp). Specifically, primerpairs for RT-PCR were 50-CTGAGATGATGGCATCTTTA-CTA-30 (KIF5B_Ex15) with 50-CTTGACCACTTTTCCAAA-TTC-30 (RET_Ex12) and 50-TGAGATTGATTCTGATGA-CACC -30 (KIF5B_Ex23) with 50- AGAGTTTTCCAAGAAC-CAAGT-30 (RET_Ex12). Total RNA (500 ng) was reverse-transcribed to cDNA using Superscript II Reverse Tran-scriptase (Invitrogen). For cDNA synthesis, we usedthe gene specific RET exon 14 reverse primer, 50-CAGG-GAGCCGTATTTGGCG-30. cDNA (corresponding to 25 ngtotal RNA) was subjected to PCR amplification usingHotStarTaq Master Mix Kit (Qiagen). The reactions werecarried out in a thermal cycler under the following con-ditions: 35 cycles of 94�C for 1 minute, 57�C for 1minute, and 72�C for 2 minutes, with a final extensionfor 7 minutes at 72�C. For sequencing, the RT-PCRproduct was cloned using TOPO TA Cloning Kit (Invitro-gen), purified with Qiaprep Spin Miniprep Kit (Qiagen)and sequenced by Sanger sequencing.

RET breakapart FISH assayA RET breakapart FISH assay was developed using

bacterial artificial chromosome (BAC) clones based onthe UCSC Genome Browser database (http://genome.ucsc.edu/). BAC clones were ordered from the Children’sHospital Oakland Research Institute (Oakland, CA). BACDNAs were extracted using BACMAX DNA purificationkit (Epicentre Biotechnologies) and labeled with eitherSpectrumOrange-dUTP (red) or SpectrumGreen-dUTP(green) using the nick-translation kit (Vysis/AbbottMolecular). RET 50-probe, a combination of BAC clonesRP11-633E1 and RP11-124O11, was labeled in green;RET 30-probe, a combination of BAC clones RP11-718J13and RP11-54P13 was labeled in red. Four-micrometerformalin-fixed, paraffin-embedded (FFPE) sections gen-erated from FFPE blocks of tumor specimens were pre-treated by deparaffinizing in xylene and dehydrating inethanol. Dual-color FISH was conducted according to theprotocol for FFPE sections from Vysis/Abbott Molecularwith a few modifications. FISH analysis and signal cap-ture were conducted on a fluorescence microscope(Zeiss) coupled with ISIS FISH Imaging System (Meta-systems). We analyzed 100 interphase nuclei from eachtumor specimen.

ResultsNanoString assay validation

To evaluate the performance of the assay design for aknown fusion, we studied 75 lung adenocarcinoma RNAsamples (6 extracted from frozen tissue; 69 from FFPEblocks) of which 24 were positive for, and 51 were negativefor, evidence of ALK fusion by ALK breakapart FISH and atleast one other method, either EML4-ALK RT-PCR or IHCusing the D5F3 ALKmonoclonal antibody (Cell Signaling).The ALK 50 to 30 expression ratios were highly nonoverlap-ping between the 2 groups, with 74 of 75 cases beingcorrectly separated (Fig. 3A). To gauge the sensitivity of theassay, we examined serial dilutions of RNAs from cell linesH2228, H3122 (both EML4-ALKþ lung adenocarcinomas)and K299 (NPM-ALKþ anaplastic large cell lymphoma),into RNA from the HL60 leukemia cell line (Fig. 3B). Weconservatively interpreted these results as suggesting thatsamples with at least 25% ALK fusion–positive tumor cellcontent should be readily detectable, an acceptable sensi-tivity range for a discovery assay. Other control samples thatwere appropriately positive included 2 ROS1 fusion–pos-itive cell lines (HCC78, U118; shown as negative controls

A

B

H3122

50% 25%

12.5%

6.25%

H2228

50%

25%

12.5%

6.25%

K299

50%

25%

U118 HCC78

100%

100%

100%

ALK

5′:3

′ ratio (

log

2)

A

LK 5

′:3′ r

atio (

log

2)

Samples 0

2

2 4 6 8 10 12 14

20 40 60

0

−2

−4

−6

−8

−10

0

−2

−4

−6

−8

−10

ALK fusion +

ALK fusion –

Figure 3. NanoString assay validation. A, detection of ALK fusions in lungadenocarcinoma samples using ALK 30:50 expression ratios. The lonediscordant case had lower RNA quality and quantity than other samples.B, serial dilutions of RNAs from cell lines with known ALK fusions. TheU118 and HCC78 cell lines are shown as negative controls. On the basisof a cutoff of log2 ratio of �4, samples with at least 25% ALK fusion–positive tumor cell content should generally be detectable. e, exon.

Suehara et al.





for ALK in Fig. 3B) and a papillary thyroid carcinomasample with a known CCDC6-RET ("RET-PTC1") fusion(not shown).

Identification of ROS1 and RET fusionsWe screened 90 tyrosine kinases and 3 RAF genes for

aberrant 50 to 30 ratios in 69 pan-negative lung adenocar-cinoma samples using this NanoString-based strategy.Examples of negative results plotted for other tyrosinekinases are shown in Supplementary Fig. S1 forAXL, FGFR1,and MET. KRAS-mutated lung adenocarcinomas, EGFR-mutated lung adenocarcinomas, and samples of non-neo-plastic lung tissue were also included as negative controls.We identified aberrant 50 to 30 ratios in ROS1 and RET in 2cases (Fig. 4), respectively, as described below.

GOPC-ROS1 fusionRACE analysis of the sample with the aberrant ROS1 50 to

30 ratio isolated a fusion of ROS1 to the Golgi-associatedPDZ and coiled-coil motif containing (GOPC) gene, previ-ously known as Fused inGlioblastoma (FIG; Fig. 5). This in-frame fusion of GOPC exon 7 to ROS1 exon 35 was alsoconfirmed by an independent RT-PCR (Fig. 5), and the RT-PCR product was also sequence-verified. This tumor wasfrom a 68-year-old white female former smoker with an 82pack-year smoking history. Histologically, the tumor was acombined adenocarcinoma and small cell carcinoma (Fig.5). Routine diagnostic IHC studies showed that the smallcell carcinoma componentwas positive for TTF-1 andCD56and had an MIB1 proliferation rate of nearly 100%. Theadenocarcinoma was weakly positive for CD56 and nega-tive for TTF-1, Napsin-A, synaptophysin, and chromogra-nin. Both components were negative for 4A4, 34BE12, andCK5/6, supporting the diagnosis. This patient’s stage IIIAlung cancer was resected in August 2010, followed byadjuvant cytotoxic chemotherapy completed by December2010, without subsequent recurrence.

KIF5B-RET fusionRACE analysis of the sample with the most aberrant RET

50 to 30 ratio isolated a fusion ofRET to the kinesin family 5Bgene (KIF5B; Fig. 6). This in-frame fusion of KIF5B exon 15to RET exon 12 was likewise also confirmed by sequencingof independent RT-PCR products. This patient was a 60-year-old female never smoker, and the tumor was a 2.8 cmadenocarcinoma with predominantly papillary and acinargrowth patterns with some solid and micropapillary com-ponents (Fig. 6). The tumor nuclei displayed frequent andprominent intranuclear inclusions. In routine diagnosticIHC studies, the tumor cells were positive for CK7, TTF-1,and PE10 and negative for CK20 and thyroglobulin, con-sistent with pulmonary origin. Using BAC clones, we devel-oped a FISH assay forRET rearrangement (seeMaterials andMethods). This showed narrow but consistent split signalsin the tumor nuclei from this case (Fig. 6), consistent withthe signal pattern of an intrachromosomal rearrangementsuch as the inversion between KIF5B and RET that wouldgenerate this gene fusion. This patient’s small, incidentallydetected adenocarcinoma was resected in March 2009without subsequent recurrence or treatment.

Further screening by RT-PCR for KIF5B-RET using mul-tiple primer combinations (see Materials andMethods) didnot identify any additional positive samples in 48 of theremaining 68 samples NanoString study set with sufficientmaterial for RT-PCR.

DiscussionThe discovery of ALK fusions in lung adenocarcinoma in

2007 resulted in renewed interest in kinase fusions asoncogenic drivers in common solid cancers (13, 14). Soonthereafter, on the basis of marked tumor shrinkagesobserved in patients with ALK rearrangements in a phaseI trial of crizotinib, a cohort of patients with ALK rearrange-ments was added and showed an overall response rate of

RE

T 5

′:3′ r

atio

(lo

g2)

Samples

U118

HCC78

Samples 0 20 40 60 100 120 14080

ROS e32 e34 e35 e40 e6

e15 e6 RET

e11 e12

5’ probe 3′ probe

3′ probe 5′ probe

NegKRASEGFRNormal

NegKRASEGFRNormal

0 20 40 60 100 120 14080

4

2

0

−2

−4

−6

4

2

0

−2

−4

−6

B

RO

S 5

′:3′ r

atio

(lo

g2)

A

Figure 4. NanoString assay results for ROS1 and RET. A and B, for bothgenes, the locationsof the 50 and30 probes are shownschematically, withthe kinase domain indicated in red. Samples with outlier negative50:30 ratios for ROS1 and RET are indicated by the arrows. Thesesampleswere subjected to 50RACE leading to the identification ofGOPC-ROS1 and KIF5B-RET fusions, respectively. The U118 and HCC78cell lines are included in the ROS1 plot as positive controls. Neg,pan-negative lung adenocarcinomas (see text); KRAS, KRAS-mutatedlung adenocarcinomas (n ¼ 17); EGFR, EGFR-mutated lungadenocarcinomas (n ¼ 11); normal, non-neoplastic lung tissue (n ¼ 37).






61% with a median response duration of 12 months, and asubsequent phase II trial of patients with ALK fusion–positive lung adenocarcinomas found a radiographicresponse rate higher than 50% (15). Together, these studiesled to the accelerated approval of crizotinib by theU.S. FoodandDrug Administration in August 2011 for this molecularsubset defined by a novel kinase fusion (16, 17). Theserecent developments prompted us to design and conduct amore systematic screen for kinase fusions in lung adeno-carcinomas without a currently known oncogenic driver,that is, "pan-negative" lung adenocarcinomas.

Using the present NanoString-based assay to screen 69"pan-negative" lung adenocarcinomas for evidence offusions involving 90 tyrosine kinase genes, we identifiednovel ROS1 and RET fusions in 2 samples. This may suggestthat undiscovered tyrosine kinase fusions are unlikely toaccount for many or most "pan-negative" lung adenocarci-nomas.However, we acknowledge that there are limitationsto the present assay design: the sensitivities for detectingoutlier 50 to 30 ratios may be variable based on the expres-sion level of the fusion gene and the expression level of thenative gene in non-neoplastic cells. Furthermore, it is likelythat additional 50 and 30 probes for each gene would allowmore robust scoring of outlier 50 to 30 ratios. As with anyhigh-throughput genomic technology, it is not possible toexperimentally validate the performance of the assay foreach of the genes tested. There are also advantages to this

NanoString-based assay, notably it extends the analyzablesample types beyond the high-quality samples usuallyrequired for array-based platforms or next-generationsequencing–based chimeric transcript detection and is,therefore, well suited to limited clinical sample RNAs(100–200 ng) of variable quality extracted from eitherfrozen tissues or FFPE tumor blocks. It is also relativelycost-efficient (approximately $200/sample).Our validationstudies and successful identification of 2 novel fusionssupport this type of assay as a discovery platform, butconsiderable additional validation studieswould beneededto establish this as a clinical diagnostic platform for one ormore tyrosine kinase fusions. Nonetheless, the labor-inten-sive, inherently low-throughput nature of FISH makesscreening for multiple possible fusion genes (ALK, ROS1,RET) cumbersome and will drive the search for moremulti-plexed diagnostic approaches such as this one or others.

We have identified for the first time the fusion of ROS1with GOPC, previously known as FIG, in a lung adenocar-cinoma. Although karyotype data were not available in ourcase, these 2 genes are only 134 KB apart in the sameorientation at 6q22 so this fusion is consistent with anintrachromosomal deletion, namely, del(6)(q22q22). TheGOPC-ROS1 fusion has previously only been found in theU118MG human glioblastoma multiforme cell line (18).The GOPC-ROS1 fusion protein has been shown to haveconstitutively active kinase activity, and its transforming

GOPC Ex7 ROS Ex35

C

c + –

A

B

Figure 5. Identification of GOPC-ROS1 fusion. A, sequence ofproduct of 50RACE shows an in-frame fusion of GOPC exon 7 toROS1 exon 35 in this cancer from a68-year-old white female with a 82pack-year smoking history. B,histology shows a combined smallcell and adenocarcinoma. Theadenocarcinoma component (leftportion of field) is poorlydifferentiated with solid and acinargrowth patterns (left inset). Thesmall cell component of the tumor(right portion of field) is composedof organoid nests and trabeculaeof densely packed cells with scantcytoplasm (right inset). The nucleiare spindle-shaped and angulatedwith finely granular chromatin andinconspicuous nucleoli (rightinset). There is marked tumornecrosis and brisk mitotic activityin both components. C, RT-PCRconfirmation of GOPC-ROS1fusion. Lane C shows the positivecontrol product (arrow) in the U188cell line RNA. Lane "þ" shows thesame product in the tumor RNA.The RT-PCR product was alsosequence-verified. Lane "�" is anegative control showing a lack ofproduct in the tumor RNAwhen theRT is omitted.

Suehara et al.





potential has been shown in a mouse transgenic modelwhere it resulted in glioblastomas in an Ink4a;Arf-null back-ground (19). ROS1 fusions in lung cancer, specificallySLC34A2-ROS1 and CD74-ROS1, were first reported byRikova and colleagues in 2007 (13). Interestingly, addition-al ROS1 fusion partners reported since then in lung cancerhave included TPM3, SDC4, EZR, and LRIG3, but notGOPC (20). All 6 lung cancer ROS1 fusions tested byTakeuchi and colleagues were found to be transforming in3T3 cells (20). Together, these data strongly support ROS1fusions as driver events, even as possible subtle functionaldifferences among the many related ROS1 fusion proteinsdescribed in lung cancer remain to be explored in lungepithelial cells or transgenicmousemodels in future studies.This molecular subset of lung adenocarcinoma is of

immediate clinical interest given the activity of crizotinib

for this target, recently shown in cell line experiments andclinically in a single ROS1 FISH–positive patient who expe-rienced a complete response to this agent (21). The expan-sion of the possible ROS1 fusion partners in lung cancer toGOPC further underscores the heterogeneity of ROS1fusions and the difficulty in using RT-PCR assays for com-prehensive detection of these patients. The size of thismolecular subset among lung adenocarcinomas was1.2% (13 of 1,116) in one study (20), 2.6% (18 of 644)in a second study (21), and 1.5% (2 of 152) in a third studylimited to Asian never smokers (22). In the present study,we found it in 1 of 69 "pan-negative" lung adenocarcino-mas, which in our testing experiencemake up about 40%oflung adenocarcinomas, and therefore our overall preva-lence of ROS1 fusions would be approximately 0.6%. Itshould be noted that certain breakapart FISH assay designs

Figure 6. Identification of KIF5B-RET fusion. A, sequence of productof 50RACE shows an in-frame fusionof KIF5B exon 15 to RET exon 12 inthis cancer from a 60-year-old neversmoker white female. B, histologyshows lung adenocarcinoma withpredominantly papillary and acinargrowth patterns (left). Nuclei arepleomorphic and display prominentintranuclear pseudoinclusions(right). C, FISH analysis showssplitting of green (50 probe) and red(30 probe) signals of RET breakapartFISH assay. Inset, normal bonemarrow cells showing fused oroverlapping red and green signalsonly.

KIF5B Ex15 RET Ex12

A

B

C






for ROS1 fusions would fail to detect the GOPC-ROS1fusion because their 50 probe overlaps or includes GOPC,which is only 134 KB upstream, and therefore no significantchange in FISH signal patterns would be apparent. This isthe case with a commercially available assay (Abnova) aswell as the assay used by Bergethon and colleagues (21).Therefore, studies based on such FISH assays may under-estimate the prevalence of ROS1 fusions by missing GOPC-ROS1 cases. In contrast, the 50 ROS1probe usedby Takeuchiand colleagues (20) includes the 50 end of ROS1 and theregion between ROS1 and GOPC, and therefore a GOPC-ROS1 fusion casemay show just a loss of the 50 ROS1 probein such an assay design due to the genomic interstitialdeletion. If the GOPC-ROS1 fusion occurs by a more com-plex mechanism such as an insertion/duplication, it maynot be detectable by any of these ROS1 FISH assays. Finally,we should note that, while the present article was underreview, another group has reported the detection of aGOPC-ROS1 fusion in lung cancer using IHC for ROS1 asa screening approach (23).

More recently, as the present report was in preparation,several independent groups (20, 24–26) also reported thediscovery of KIF5B-RET fusions in lung adenocarcinomas,at anoverall prevalence of approximately 1.3% (27).On thebasis of the same estimates provided above for ROS1fusions, our overall prevalence of RET fusions would beapproximately 0.6%. Notably, our KIF5B-RET–positivetumor was detected in 1 of only 5 never smokers in ourdiscovery set, suggesting that screening of "pan-negative"never smokers may be a more efficient strategy to identifythis rather small subset of patients. It is possible that morecases with kinase fusions might have been detected if ourstudy group had been intentionally enriched for neversmokers. Our KIF5B-RET fusion involved exon 15 of KIF5Band exon 12 of RET, thus retaining a portion of the dimer-ization domain of KIF5B and the entire kinase domain ofRET, analogous to RET fusion proteins in papillary thyroidcarcinoma. This fusion structure is also the most com-monly reported of the seven isoforms described so far(21). Cytogenetically, as KIF5B and RET are approximately10.6Mb apart on chromosome 10 in opposite orientations,the KIF5B-RET fusion may occur most simply by a peri-centric inversion of 10p11.22-q11.21. RET is a well-estab-lished oncogene in thyroid cancer, both in the context ofpoint mutations in medullary thyroid cancer and otherfusions in papillary thyroid cancer (28). The oncogenicpotential of KIF5B-RET itself has been established by trans-fection of NIH 3T3 cells in which it leads to anchorage-independent growth in vitro and tumor formation in xeno-grafted mice (20, 25) and by transfection of Ba/F3 cells inwhich it results in interleukin-3–independent growth(20, 24). Although no human lung cancer cell line with anendogenous KIF5B-RET fusion has been identified, thephenotypes elicited in heterologous cells by transfectionwith KIF5B-RET cDNA are sensitive to multi-kinase inhibi-tors whose targets include RET, such as vandetanib, suni-tinib, and sorafenib (20, 24, 25), supporting the concept ofprospectively genotyping lung cancers for RET fusions to

identify patients for clinical trials of such agents in this newmolecular subset of lung adenocarcinoma. It is interestingto note that agents such as sunitinib (29, 30), sorafenib(31), and cabozantinib (or XL184) in unselected patientswith non–small cell lung cancer have been associated withpartial response rates in the 2% to 10% range. Given themulti-kinase targeting of these agents, it has been difficult toelucidate why some patients benefit but it is now temptingto speculate that some of these partial responses may havebeen in tumors containing a RET fusion.

Finally, it is notable that the 2 most common lungfusions so far, EML4-ALK and KIF5B-RET, are both intra-chromosomal events, as is the only other RET fusiondescribed so far in lung cancer, CCDC6-RET, reported inone case by Takeuchi and colleagues (20). Although mostROS1 fusion partners are on other chromosomes, theGOPC-ROS1 fusion reported here and another recentlydescribed ROS1 fusion, EZR-ROS1 (20), are also intra-chromosomal rearrangements. These observations areintriguing given the link between radiation and intra-chromosomal rearrangements established in papillarythyroid cancer based on epidemiologic and mechanisticdata. Specifically, papillary thyroid cancers that developfollowing nuclear disasters are more likely to harbor RETrearrangements (instead of BRAF or RAS mutations;refs. 32, 33) and the fusion partners are more oftenCCDC6 ("RET-PTC1" fusion) and NCOA4 ("RET-PTC3"fusion), both on chromosome 10 (32). RET fusions canbe induced in vitro by irradiating human thyroid epithe-lial cells and these 3-chromosome 10 loci (RET, CCDC6,NCOA4) are often juxtaposed in the interphase nuclei ofthyroid epithelial cells, which may facilitate coincidentDNA strand breakage by a single radiation track and/orillegitimate recombination (34–36). These considera-tions, along with the increased risk of lung cancer inatomic bomb survivors and uranium miners (37, 38),suggest that a possible etiologic role for environmentalradiation should be explored in never smokers whosetumors harbor KIF5B-RET or possibly EML4-ALK.

Disclosure of Potential Conflicts of InterestM.G. Kris has a commercial research grant from Pfizer, Inc. and

Boehringer Ingelheim and is a consultant/advisory board member forPfizer, Inc. M. Ladanyi has been a consultant/advisory board member forNanoString Technologies, Inc. No potential conflicts of interest weredisclosed by the other authors.

Authors' ContributionsConception and design: M. Arcila, M. LadanyiDevelopment of methodology: Y. Suehara, L. Wang, D. Ang, M. LadanyiAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): Y. Suehara, M. Arcila, L. Wang, A. Hasanovic, D.Ang, A. Drilon, U. Guha, V. Rusch, M.G. Kris, M.F. Zakowski, N. RizviAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis):Y. Suehara,M.Arcila, Y. Kimura,U.Guha,M.G. Kris, M.F. Zakowski, R. Khanin, M. LadanyiWriting, review, and/or revisionof themanuscript: Y. Suehara,M. Arcila,L. Wang, A. Hasanovic, D. Ang, Y. Kimura, A. Drilon, V. Rusch, M.G. Kris, N.Rizvi, M. LadanyiAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): Y. Suehara, A. Hasanovic, T. Ito, Y.Kimura, M.G. Kris, M. LadanyiStudy supervision: Y. Kimura, M.G. Kris, M. Ladanyi

Suehara et al.





Developed FISH assays to verify KIF5B-RET and GOPC-ROS fusionsidentified by nanostring and RACE and to screen for these fusions inadditional samples: L. Wang

AcknowledgmentsThe authors thank William Pao and Juliann Chmielecki for sharing

lung cancer cell lines and bioinformatic data on tyrosine kinase genes,James Fagin for advice and providing RNA from a CCDC6-RET–positivethyroid cancer, Laetitia Borsu and Adriana Heguy for assistance with theNanoString platform, and Cameron Brennan for providing U118 cell lineRNA.

Grant SupportThe study was supported byNIH P01CA129243 (toM. Ladanyi andM.G.

Kris), Uniting Against Lung Cancer (to M. Ladanyi), and an InternationalAssociation for the Study of Lung Cancer Fellowship (to A. Drilon).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received April 10, 2012; revised August 30, 2012; accepted September 27,2012; published OnlineFirst October 10, 2012.

References1. Pao W, Iafrate AJ, Su Z. Genetically informed lung cancer medicine.

J Pathol 2011;223:230–40.2. Pao W, Girard N. New driver mutations in non-small-cell lung cancer.

Lancet Oncol 2011;12:175–80.3. Johnson BE, Kris MG, Kwiatkowski DJ, Iafrate AJ, Wistuba II, Shyr Y,

et al. Clinical characteristics of planned 1000 patients with adenocar-cinoma of lung (ACL) undergoing genomic characterization in the USLung Cancer Mutation Consortium (LCMC). J Thorac Oncol 2011;6:S346.

4. D'Angelo SP, Park B, Azzoli CG, Kris MG, Rusch V, Ladanyi M, et al.Reflex testing of resected stage I through III lung adenocarcinomas forEGFR and KRAS mutation: report on initial experience and clinicalutility at a single center. J Thor Cardiovasc Surg 2011;141:476–80.

5. Arcila ME, Chaft JE, Nafa K, Roy-Chowdhuri S, Lau C, Zaidinski M,et al. Prevalence, clinicopathologic associations and molecular spec-trum of ERBB2 (HER2) tyrosine kinase mutations in lung adenocarci-nomas. Clin Cancer Res 2012;18:4910–8.

6. Wang L, Motoi T, Khanin R, Olshen A, Mertens F, Bridge J, et al.Identification of a novel, recurrent HEY1-NCOA2 fusion in mesenchy-mal chondrosarcoma based on a genome-wide screen of exon-levelexpression data. Genes Chromosomes Cancer 2012;51:127–39.

7. Su Z, Dias-Santagata D, Duke M, Hutchinson K, Lin YL, Borger DR,et al. A platform for rapid detection of multiple oncogenic mutationswith relevance to targeted therapy in non-small-cell lung cancer. J MolDiagn 2011;13:74–84.

8. Pan Q, Pao W, Ladanyi M. Rapid PCR-based detection of epidermalgrowth factor receptor genemutations in lungadenocarcinomas. JMolDiagn 2005;7:396–403.

9. Chaft JE, Arcila ME, Paik PK, Lau C, Riely GJ, Pietanza MC, et al.Coexistence of PIK3CA and other oncogene mutations in lung ade-nocarcinoma-rationale for comprehensive mutation profiling. MolCancer Ther 2012;11:485–91.

10. Reis PP, Waldron L, Goswami RS, Xu W, Xuan Y, Perez-Ordonez B,et al. mRNA transcript quantification in archival samples using multi-plexed, color-coded probes. BMC Biotechnol 2011;11:46.

11. Payton JE, Grieselhuber NR, Chang LW, Murakami M, Geiss GK, LinkDC, et al. High throughput digital quantification ofmRNAabundance inprimary human acute myeloid leukemia samples. J Clin Invest2009;119:1714–26.

12. Chmielecki J, Peifer M, Jia P, Socci ND, Hutchinson K, Viale A, et al.Targeted next-generation sequencing of DNA regions proximal to aconserved GXGXXG signaling motif enables systematic discoveryof tyrosine kinase fusions in cancer. Nucleic Acids Res 2010;38:6985–96.

13. Rikova K, Guo A, Zeng Q, Possemato A, Yu J, Haack H, et al. Globalsurvey of phosphotyrosine signaling identifies oncogenic kinases inlung cancer. Cell 2007;131:1190–203.

14. Soda M, Choi YL, Enomoto M, Takada S, Yamashita Y, Ishikawa S,et al. Identification of the transforming EML4-ALK fusion gene in non-small-cell lung cancer. Nature 2007;448:561–6.

15. Kwak EL, Bang YJ, Camidge DR, ShawAT, Solomon B,Maki RG, et al.Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer. NEngl J Med 2010;363:1693–703.

16. Riely GJ, Chaft JE, Ladanyi M, Kris MG. Incorporation of crizotinib intothe NCCN guidelines. J Natl Compr Canc Netw 2011;9:1328–30.

17. Gerber DE, Minna JD. ALK inhibition for non-small cell lung cancer:from discovery to therapy in record time. Cancer Cell 2010;18:548–51.

18. Charest A, Lane K, McMahon K, Park J, Preisinger E, Conroy H, et al.Fusion of FIG to the receptor tyrosine kinase ROS1 in a glioblastomawith an interstitial del(6)(q21q21). Genes Chromosomes Cancer2003;37:58–71.

19. Charest A, Wilker EW, McLaughlin ME, Lane K, Gowda R, Coven S,et al. ROS1 fusion tyrosine kinase activates a SH2 domain-containingphosphatase-2/phosphatidylinositol 3-kinase/mammalian target ofrapamycin signaling axis to form glioblastoma in mice. Cancer Res2006;66:7473–81.

20. Takeuchi K, Soda M, Togashi Y, Suzuki R, Sakata S, Hatano S, et al.RET, ROS1 and ALK fusions in lung cancer. Nat Med 2012;18:378–81.

21. Bergethon K, Shaw AT, Ignatius Ou SH, Katayama R, Lovly CM,McDonald NT, et al. ROS1 rearrangements define a unique molecularclass of lung cancers. J Clin Oncol 2012;30:863–70.

22. Li C, Fang R, Sun Y, Han X, Li F, Gao B, et al. Spectrum of oncogenicdriver mutations in lung adenocarcinomas from East Asian neversmokers. PLoS One 2011;6:e28204.

23. RimkunasVM,Crosby KE, Li D, HuY, KellyME, GuTL, et al. Analysis ofreceptor tyrosine kinase ROS1-positive tumors in non-small cell lungcancer: identification of a FIG-ROS1 fusion. Clin Cancer Res 2012;18:4449–57.

24. Lipson D, Capelletti M, Yelensky R, Otto G, Parker A, Jarosz M, et al.Identification of new ALK and RET gene fusions from colorectal andlung cancer biopsies. Nat Med 2012;18:382–4.

25. Kohno T, Ichikawa H, Totoki Y, Yasuda K, Hiramoto M, Nammo T,et al. KIF5B-RET fusions in lung adenocarcinoma. Nat Med 2012;18:375–7.

26. Ju YS, Lee WC, Shin JY, Lee S, Bleazard T, Won JK, et al. Atransforming KIF5B and RET gene fusion in lung adenocarcinomarevealed fromwhole-genomeand transcriptome sequencing.GenomeRes 2012;22:436–45.

27. PaoW,HutchinsonKE. Chipping away at the lung cancer genome. NatMed 2012;18:349–51.

28. Castellone MD, Santoro M. Dysregulated RET signaling in thyroidcancer. Endocrinol Metab Clin North Am 2008;37:363–74, viii.

29. SocinskiMA,Novello S, Brahmer JR,Rosell R, Sanchez JM,Belani CP,et al. Multicenter, phase II trial of sunitinib in previously treated,advanced non-small-cell lung cancer. J Clin Oncol 2008;26:650–6.

30. Novello S, Scagliotti GV, Rosell R, Socinski MA, Brahmer J, Atkins J,et al. Phase II study of continuous daily sunitinib dosing in patientswithpreviously treated advanced non-small cell lung cancer. Br J Cancer2009;101:1543–8.

31. Scagliotti G, Novello S, von PJ, Reck M, Pereira JR, Thomas M,et al. Phase III study of carboplatin and paclitaxel alone or withsorafenib in advanced non-small-cell lung cancer. J Clin Oncol2010;28:1835–42.

32. Nikiforov YE, Rowland JM, Bove KE, Monforte-Munoz H, Fagin JA.Distinct pattern of ret oncogene rearrangements in morphologicalvariants of radiation-induced and sporadic thyroid papillary carcino-mas in children. Cancer Res 1997;57:1690–4.

33. Hamatani K, Eguchi H, Ito R,MukaiM, Takahashi K, TagaM, et al. RET/PTC rearrangements preferentially occurred in papillary thyroid cancer






among atomic bomb survivors exposed to high radiation dose. CancerRes 2008;68:7176–82.

34. Nikiforov YE, Koshoffer A, Nikiforova M, Stringer J, Fagin JA.Chromosomal breakpoint positions suggest a direct role for radi-ation in inducing illegitimate recombination between the ELE1 andRET genes in radiation-induced thyroid carcinomas. Oncogene1999;18:6330–4.

35. Nikiforova MN, Stringer JR, Blough R, Medvedovic M, Fagin JA,Nikiforov YE. Proximity of chromosomal loci that participate in radi-ation-induced rearrangements in human cells. Science 2000;290:138–41.

36. Gandhi M, Evdokimova V, Nikiforov YE. Mechanisms of chromosomalrearrangements in solid tumors: the model of papillary thyroid carci-noma. Mol Cell Endocrinol 2010;321:36–43.

37. Ozasa K, Shimizu Y, Suyama A, Kasagi F, Soda M, Grant EJ, et al.Studies of the mortality of atomic bomb survivors, report 14, 1950–2003: an overview of cancer and noncancer diseases. Radiat Res2012;177:229–43.

38. Rage E, Vacquier B, Blanchardon E, Allodji RS, Marsh JW, Caer-LorhoS, et al. Risk of lung cancer mortality in relation to lung doses amongFrench uranium miners: follow-up 1956–1999. Radiat Res 2012;177:288–97.

Suehara et al.





2012;18:6599-6608. Published OnlineFirst October 10, 2012.Clin Cancer Res Yoshiyuki Suehara, Maria Arcila, Lu Wang, et al. for Tyrosine Kinase FusionsAdenocarcinomas through a Comprehensive mRNA-Based Screen

Fusions in LungGOPC-ROS1 and KIF5B-RETIdentification of

Updated version

10.1158/1078-0432.CCR-12-0838doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2012/10/10/1078-0432.CCR-12-0838.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/18/24/6599.full#ref-list-1

This article cites 38 articles, 12 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/18/24/6599.full#related-urls

This article has been cited by 20 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/18/24/6599To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-12-0838

http://clincancerres.aacrjournals.org/content/suppl/2012/10/10/1078-0432.CCR-12-0838.DC1

http://clincancerres.aacrjournals.org/content/18/24/6599.full#ref-list-1

http://clincancerres.aacrjournals.org/content/18/24/6599.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/18/24/6599


Identiﬁcationof KIF5B-RET and GOPC-ROS1 FusionsinLung ... · the nononcogenic, reciprocal fusion...

Documents

Transcript of Identiﬁcationof KIF5B-RET and GOPC-ROS1 FusionsinLung ... · the nononcogenic, reciprocal fusion...