the fa/brca pathway identiﬁed as the major predictor of ......companion diagnostics and cancer...

Companion Diagnostics and Cancer Biomarkers

The FA/BRCA Pathway Identified as the MajorPredictor of Cisplatin Response in Head and NeckCancer by Functional GenomicsSanne R. Martens-de Kemp1, Arjen Brink1, Ida H. van der Meulen2, Ren�ee X. de Menezes3,Dennis E. te Beest3, C. Ren�e Leemans1, Victor W. van Beusechem2,Boudewijn J.M. Braakhuis1, and Ruud H. Brakenhoff1

Abstract

Patients with advanced stage head and neck squamous cellcarcinoma (HNSCC) are often treated with cisplatin-containingchemoradiation protocols. Although cisplatin is an effective radi-ation sensitizer, it causes severe toxicity andnot all patients benefitfrom the combination treatment. HNSCCs expectedly notresponding to cisplatin may better be treated with surgery andpostoperative radiation or cetuximab and radiation, but biomar-kers to personalize chemoradiotherapy are not available. Weperformed an unbiased genome-wide functional genetic screenin vitro to identify genes that influence the response to cisplatin inHNSCC cells. By siRNA-mediated knockdown, we identified the

Fanconi anemia/BRCA pathway as the predominant pathway forcisplatin response inHNSCC cells. We also identified the involve-ment of the SHFM1 gene in the process of DNA cross-link repair.Furthermore, expression profiles based on these genes predict theprognosis of radiation- and chemoradiation-treated head andneck cancer patients. This genome-wide functional analysis des-ignated the genes that are important in the response of HNSCC tocisplatin and may guide further biomarker validation. Cisplatinimaging as well as biomarkers that indicate the activity of theFanconi anemia/BRCA pathway in the tumors are the primecandidates. Mol Cancer Ther; 16(3); 540–50. �2016 AACR.

IntroductionCancer of the head and neck is the eighth most commonly

diagnosed cancer worldwide (1). Over 90% of the head and neckcancers are squamous cell carcinomas that arise in the mucosallinings of the upper aerodigestive tract. Approximately 60% ofhead and neck squamous cell carcinoma (HNSCC) patientspresent with advanced stage disease (stage III and IV), whichrelates to a poor prognosis (2). Treatment options for this groupare surgery followed by radiotherapy or cisplatin-containingchemoradiation protocols with salvage surgery if needed.

Although chemoradiation is effective (3), not all tumorsrespond well to this combination of cisplatin and radiotherapy.Cisplatin is an effective and inexpensive addendum to radiationprotocols, but only between 5% and 10% of the patients benefitfrom the combination at the expense of severe toxicity. Patientsneed to be hospitalized during cisplatin infusion and acute andlong term toxicities are frequent. It would be ideal to personalize

the application of chemoradiation and to treat only those patientswhose tumors are likely to respond to the combination therapy.

Major research efforts aimed to find clinical and biomolec-ular markers that predict chemoradiation response of HNSCC.However, the only clinical factor that shows some predictivevalue for chemoradiotherapy outcome in HNSCC proved to beprimary tumor volume (4–11), and not even all studies couldconfirm this (12).

Using a candidate gene or protein approach, several biologicaland genetic markers have been studied to predict chemoradiationoutcome in head and neck cancer patients but none found itsway to the clinic (13–20).Other groups have exploitedmicroarraytechnology to determine expression profiles that might predicttreatment response in head and neck cancer (12, 21–28), but apredictive profile has not been validated.

There are several explanations why these approaches have notled tobreakthroughs. First, tumorbiopsiesmay containmixed cellpopulations, and particularly the small subpopulations of treat-ment-resistant cells (e.g., cancer stem cells) may not be analyzedby expression profiling, while they might be highly relevant fortreatment outcome (29). Second, genes harboring an activating orinactivating mutation, but that are expressed at close to normallevels, will not be identified as predictors of treatment outcome.

We therefore aimed to identify all genes that have an importantrole in the response to cisplatin using a functional genomicsapproach. Unveiling these genes might enable us to find biomar-kers that can be used to predict cisplatin-based chemoradiationoutcome and to personalize HNSCC treatment by only treatingthose patients who would benefit from the therapy.

The application of loss-of-function high-throughput RNAinterference screens has become an important genomic tool in

1Department of Otolaryngology-Head and Neck Surgery, VU University MedicalCenter, Amsterdam, the Netherlands. 2RNA Interference Functional Oncoge-nomics Laboratory, Department of Medical Oncology, VU University MedicalCenter, Amsterdam, the Netherlands. 3Department of Clinical Epidemiology andBiostatistics, VU University Medical Center, Amsterdam, The Netherlands.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Ruud H. Brakenhoff, VU University Medical Center, POBox 7057, Amsterdam 1007 MB, the Netherlands. Phone: 312-0444-0953; Fax:312-0444-3688; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-16-0457

�2016 American Association for Cancer Research.

MolecularCancerTherapeutics

Mol Cancer Ther; 16(3) March 2017540

on July 21, 2021. © 2017 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst December 15, 2016; DOI: 10.1158/1535-7163.MCT-16-0457

http://mct.aacrjournals.org/

research (30–33), although they are rapidly takenover byCRISPR-Cas9–mediated gene editing approaches. Especially, for the iden-tification of genes that modulate drug response, screens usingsiRNAs are increasingly used (34–43). For example, using thistechnology, Posthuma De Boer and colleagues (44) identifiedJIP1 as a gene that modulates the sensitivity of osteosarcoma cellsto doxorubicin. We hypothesized that a genome-wide siRNAscreen could be the best way to functionally identify all genesand signaling pathways involved in cisplatin response. Therefore,we employed an arrayed library of 21,121 pools of siRNAs,targeting unique human genes in the NCBI RefSeq database(www.ncbi.nlm.nih.gov/RefSeq and ref. 45). This one-gene-one-well approach allowed us to comprehensively interrogatethe cellular response to cisplatin when expression of each indi-vidual gene is knocked down separately. Resultsmay lead to novelbiomarkers that predict which HNSCC patients will benefit fromcisplatin-containing protocols.

Materials and MethodsCell lines, clinical specimens, and chemicals

All cancer cell lines were cultured inDMEM(Lonza) containing5% FCS (Lonza) and 2 mmol/L L-glutamine (Lonza). Cells weregrown in a humidified atmosphere of 5% CO2 at 37�C.

HNSCC cell lines UM-SCC-11B (a larynx tumor; T2N2A) andUM-SCC-22B (a hypopharynx tumor; T2N1) were obtained fromDr. T. Carey (University of Michigan, Ann Arbor, MI) in the year1984 (46). Cell line VU-SCC-120 (formerly known as 93VU120)was established in 1993, as described previously (47). All HNSCCcell lines were authenticated to the earliest available passages bymicrosatellite profiling and TP53 mutation analysis (46, 48).

Clinical specimensof apanel of 22HNSCCsand correspondingmacroscopically normal mucosa, adjacent to the tumor, werecollected according to the Dutch legislations on research withhuman material. This study was approved by the InstitutionalReview Board VU University Medical Center (VUmc; Amsterdam,the Netherlands). Collection of specimens was done immediatelyafter surgery, prior to radiotherapy, in the period 2000–2007.Samples were snap frozen and stored in liquid nitrogen. Inclusioncriteria were squamous cell carcinoma from the oral cavity ororopharynx, and the treatment consisted of surgery with orwithout postoperative radiotherapy. All normal mucosa sampleswere judged free from dysplasia by an experienced pathologist.The panel included 13males and 9 females with an average age of60.0� 10.1 years (range 33.8–82.5). The majority of the patientshad a history of heavy smoking (�30 pack-years; 17/22, 77%)and/or heavy alcohol consumption (�100 unit-years, with 1 unitis equivalent to approximately 15 mL of ethanol; 15/22, 68%).Specimens were obtained from the oral cavity (17/22, 77%) andthe oropharynx (5/22, 23%) with the majority showing a mod-erate degree of differentiation (15/22, 68%), and the remainingseven cases were poorly differentiated. All orophayngeal tumorswere tested as human papillomavirus negative. Twenty-threepercent of the patients had early disease stage and 77% advancedstage. Cisplatin was obtained from Teva Pharmachemie in aconcentration of 1 mg/mL.

IC50 measurementsCells were plated in 96-well plates and were treated with 18

different concentrations of cisplatin (ranging from 0.01 to 666mmol/L) 24 hours later. Plates were incubated for 72 hours at

37�C/5% CO2. Cell viability was calculated on the basis offluorescence measurements after 2 hours incubation with 20 mLof a 1:1 dilution of CellTiter Blue (Promega) in cell culturemedium. Fluorescence wasmeasured using an Infinite 200micro-plate reader (Tecan).

Genome-wide siRNA screenVU-SCC-120 cells were subjected to high-throughput forward

transfection as described before (49). In summary, 1,000 cellswere seeded and the next day transfected with 25 nmol of eachsiRNA SMARTpool derived from the siARRAY Human Genomelibrary [Catalog items G-003500 (Sept05), G-003600 (Sept05),G-004600 (Sept05), and G-005000 (Oct05); Dharmacon,Thermo Fisher Scientific] and 0.03 mL DharmaFECT1 (ThermoFisher Scientific). The nontargeting siCONTROL#2 and the PLK1targeting siRNA SMARTpool were used as negative and positivecontrol for transfection efficiency, respectively. After 24 hoursof incubationwith the transfection reagents,mediumvolumewasreduced to 40 mL using theMicrolab STAR liquid handling station(Hamilton Robotics), and 60 mL freshmediumwas added with orwithout cisplatin to reach final concentrations matching IC0 (nocisplatin) or IC20 (2 mmol/L) using a Multidrop Combi (ThermoFisher Scientific). Plates were incubated for 72 hours at 37�C/5%CO2. Cell viability was measured using CellTiter Blue (Promega)as described.

Deconvolution of siRNA SMARTpools that significantly causedcellular sensitivity to cisplatin was performed using the sameprocedures, either by hand or using the same automatedmethodsas described above. Deconvolution of the same siRNA SMART-pools was investigated in two additional HNSCC cell lines, UM-SCC-11B and UM-SCC-22B. These were chosen based on theirsensitivity to cisplatin, having IC20 values that are two times lowerand four times lower than for VU-SCC-120, respectively. Inaddition, both cell lines could be reproducibly transfected usingthe robotic platform described previously.

UM-SCC-22B was transfected with 25 nmol siRNA and 0.10 mLDharmaFECT1. UM-SCC-11Bwas transfected with 25 nmol siRNAand0.065mLDharmaFECT1. IC20 cisplatinvalues forUM-SCC-11Band UM-SCC-22B were 1.0 and 0.05 mmol/L, respectively.

Data analysis of the genome-wide screenThe data from our genome-wide high-throughput screen were

analyzed with two different statistical methods; Z-score calcula-tion and a Limma-based linear regression model.

For the Z-score analysis, raw fluorescence values were log10transformed and per plate the trimmed mean was calculated(trimming factor 0.5). Plate normalization was accomplishedusing the overall mean per cisplatin condition (IC0 or IC20) andper replicate. The normalized fluorescent values were back-trans-formed. Per cisplatin condition, both replicates were normalizedtoward each other and for each siRNA replicates of each cisplatinconditionwere averaged. IC0/IC20 ratios were calculated and usedfor Z-score calculations.

For the calculations in Limma, the data were read into R(version 2.12) and configured using cellHTS2 (50). Subsequently,the data were log2 transformed and normalized by correcting forlinear effects of "day" and "plate" using a regression model.To estimate the treatment effect, an empirical Bayes linear modelaccounting for the treatment level (51) was fitted to the normal-ized data using the BioConductor package Limma (52). Theresulting P value list was subsequently corrected for multiple

The FA/BRCA Pathway Predicts Cisplatin Response in HNSCC

www.aacrjournals.org Mol Cancer Ther; 16(3) March 2017 541




testing using Benjamini–Hochberg's step-up FDR method (53).This approach yields more power to find associations, in partic-ular in studies with a small sample size as the current one (54).

Deconvolution of siRNA poolsGene-specific knockdown was confirmed by investigation of

the phenotype caused by each of the four separate siRNAs thatmake up the siRNA pool. Triplicate experiments were performedas described above and themedian of the three IC0measurementsas well as the median of the three IC20 measurements werecalculated. With these medians, P values against siCONTROL#2measurements were obtained with the Student t test. We previ-ously analyzed over 1,000 siCONTROL#2 measurements andobserved that the data follow the normal distribution (Supple-mentary Fig. S1), allowing t test analysis.

Quantitative real-time PCRCells were seeded in 96-well plates and transfected as described

above. RNA was extracted 96 hours post-transfection using theRNeasyMicroKit (Qiagen). cDNAwas synthesized from250ngofRNA using AMV reverse transcriptase (Promega) and a customdesigned reverse primer (Supplementary Table S1). Amplificationof the cDNA was performed on the ABI/Prism 7500 SequenceDetector System (TaqMan-PCR; Applied Biosystems) with 1�Power SYBR Green PCR Master Mix (Applied Biosystems) andcustomdesignedprimers for eachgeneof interest (SupplementaryTable S1). CTR1 expression after siRNA transfection was investi-gated using TaqMan Gene Expression Assay (Hs00977268_g1;AppliedBiosystems) according to themanufacturer's instructions.

For each sample, the cycle number at which the amount ofamplified target crossed a predetermined threshold (the Ct value)was determined. To correct for differences in RNA input andquality, b-glucuronidase (GUSB) was used as a housekeepinggene (55) for each RNA sample. The mRNA expression wascalculated relative to that of GUSB using the DDCt method(56). Quantitative real-time (RT)-PCR reactions without reversetranscriptase were carried out in parallel for each RNA sample toexclude signal from contaminating genomic DNA.

Cell-cycle analysisCells were seeded in 25-cm2

flasks and transfected 24 hourslater in duplicate with siCONTROL#2, BRCA2 SMARTpool, orSHFM1 SMARTpool. After another 24 hours, the cultures wererefreshed with either fresh medium or medium containing 500nmol/L cisplatin. After 72 hours of incubation, the cells wereharvested and fixed with 70% EtOH during at least 3 hours. Next,cells were incubated with 0.5 mg/mL RNAse A in PBS at 37�C for30 minutes. The cells were washed and the DNA content wasstained with propidium iodide (PI). The cell-cycle distributionwas analyzed with a BD FACSCalibur flow cytometer (BD Bios-ciences). Cell-cycle analyses were performed using BD CellQuestsoftware (BD Biosciences) and calculations in Microsoft Excel.

Expression microarrayRNA was isolated with TRIzol (Life Technologies) according to

the manufacturer's instructions. RNA integrity was measuredusing a RNA nanochip on an Agilent 2100 Bioanalyzer (AgilentTechnologies). Synthesis and labeling of cDNA was performedaccording to the recommendations of themanufacturer (Agilent),including quality controls. The mRNA expression profiles were

determined using a mRNA microarray (4 � 44 K Whole HumanGenome Arrays G4112F; Agilent). Labeling, hybridization, andscanning was performed according to the manufacturer's proto-col, using aG2505B scanner and Feature Extraction v9.5 (Agilent).Data analysis was performed as described previously (57) and canbe downloaded using GEO accession number GSE83519.

ResultsGenome-wide siRNA screen design

Nineteen HNSCC cell lines were subjected to a selection pro-cedure to identify the cell line most suitable for genome-widesiRNA application (data not shown). Selection criteria were basedon cisplatin sensitivity, growth rate at low density in 96-wellplates, limited acidification of themedium to allowCellTiter-Blueassays, and reproducibility of siRNA transfection, also on arobotic platform. Cell line VU-SCC-120, a cell line derived froma previously untreated moderately differentiated, advanced stagetongue cancer (T3N1), matched all criteria. The half-maximalinhibitory concentration (IC50) of cisplatin was determined at 4.0mmol/L for this cell line (58). Because we are interested in siRNAsthat increase the response to cisplatin, we decided to screen thesiRNA library in the presence of a cisplatin concentration thataffected cell survival by 20% (IC20), resulting in a theoreticalscreening window of 80%. For VU-SCC-120, this meant that thecells were treated with 2.0 mmol/L of cisplatin (Fig. 1A).

We used these data to design a robust, high-throughput RNAinterference screen for the VU-SCC-120 cell line. Cells were trans-fected in quadruplicate in separate 96-well plates for duplicateanalysis of IC0 (no cisplatin) and IC20 (Fig. 1B). We used the PLK1siRNA SMARTpool, which resulted in a decrease in cell viability(data not shown), as expected (59), as a positive control fortransfection efficiency. The nontargeting siCONTROL#2 was usedas negative control. Cell viability was measured in every plate toevaluate the effect of each siRNAon cell growth (IC0) and the effectof reduced gene expression on cisplatin sensitivity (IC0 vs. IC20).

Genome-wide siRNA screen to identify cisplatin susceptibilitygenes

Raw cell viability values were inspected for transfection-induced cell death (siCONTROL#2 vs. untransfected cells; 16%cell death) and transfection efficiency (siPLK1 vs. siCONTROL#2;92% cell death), and were found to be acceptable (Z0 factor ¼0.570). Comparison of the two independent duplicate screensshowed high reproducibility with Spearman r ¼ 0.823. Subse-quently, the data were analyzed using two independent statisticalmethods as described in the Materials and Methods; Z-scoreanalysis and a Limma-based linear regression model. Z-scorecalculations yielded 205 siRNA pools that significantly sensitizedVU-SCC-120 cells to cisplatin (Z � �3; P < 0.001; Fig. 1C),whereas the linear regression model resulted in 168 hits (rawP < 0.05). The hit lists obtained with these two methods shared102 siRNAs (Fig. 1D; Supplementary Table S2).

We subsequently selected siRNAs for stratification by decon-volution of the siRNA pools into four separate siRNAs. From the102 siRNA pools that were found to cause a significant sensiti-zation to cisplatin by both statistical analyses, we were able toorder the individual siRNAs of 97 pools. The target sequences ofthe remaining five siRNA pools were no longer annotated aspotential genes. In addition, we randomly selected 28 siRNAsthat were only identified byZ-score analysis and 23 that were only

Martens-de Kemp et al.

Mol Cancer Ther; 16(3) March 2017 Molecular Cancer Therapeutics542




found by the linear regression model for further analysis. Decon-volution was done in VU-SCC-120, the cell line used for thegenome-wide siRNA discovery screen, and two additionalHNSCC cell lines, UM-SCC-11B and UM-SCC-22B.

We marked an siRNA as "confirmed" in case at least twoseparate siRNAs significantly (P < 0.05, Student t test againstsiCONTROL#2 transfections) sensitized the tumor cells to cis-platin (Supplementary Table S3). We could confirm 21% of thehits identified by only the Z-score analysis and 17% of the hitsidentified by the linear regression model only. However, theconfirmation rate obtained by siRNAs that were identified bybothmethods proved to be superior (42% confirmed) over eithermethod alone. This observation also holds true for thenumbers ofsiRNAs that were confirmed in at least two cell lines or even in allthree HNSCC cell lines. In total, 51 of the tested 148 primary hitswere confirmed in VU-SCC-120 cells, of which 21 were repro-ducible in all three HNSSC cell lines (Table 1).

The FA/BRCA pathwayComprehensive pathway analysis on the complete data set by

IPA (Ingenuity systems) revealed that particularly genes in theFanconi anemia/BRCA pathway, which is involved in DNA cross-link repair, are implicated in cisplatin response in HNSCC cells

(P¼ 2.13� 10�4; Fig. 2A). In addition, according to the STRINGdatabase (version 9.05), the only cluster that could be foundamong the siRNAs that were confirmed in all three HNSCC celllines (n¼ 21; Table 1), harbored theseDNA repair genes (Fig. 2B).This indicates that an important determinant for cisplatin sensi-tivity in HNSCC cells is the proper functioning of particularly theFA/BRCA pathway.

Indeed, close examination of two important players in theFA/BRCA pathway, BRCA1 and BRCA2, showed that knockdownof these genes enhances cisplatin-induced cell death (Fig. 2C andSupplementary Fig. S2A). Also the split hand/foot malformationtype 1 gene (SHFM1 or DSS1), which has been shown to phys-ically interact with BRCA2 (60–63), could be confirmed as adeterminant of cisplatin response with all four separate siRNAs(Fig. 2C), and this was confirmed in multiple cell lines (Supple-mentary Fig. S2B). Knockdown of mRNA expression by theseparate SHFM1 siRNAs was determined by quantitative RT-PCRand showed to correlate with the increase in cisplatin-induced celldeath (Supplementary Fig. S2A), demonstrating that these arespecific effects caused by knockdown of the SHFM1 gene and notdue to off-target effects.

Because SHFM1 is known to interact with BRCA2, we ques-tioned whether knockdown of SHFM1 would cause a cell-cycle

Day 1Seed cells

4×

Day 2Transfection

4×

Day 3Drug treatment

Day 6Read out

Sens

itivi

ty

2× Mock (IC0)

2× Cisplatin (IC20)10-2 10-1 100 101 102 103

0

50

100

Cisplatin (μmol/L)

IC20

Surv

ivin

g po

pula

tion

(%)

IC50

A B

C

siRNA poolsiCONTROL#2

Z-sc

ore

6

0

−3

−9

−15

0 5000 10000 15000 20000

3

−12

−6

D

Linear regressionmodel

(raw P < 0.05)

Z-score analysis

(Z≥-3)103 102 66

Figure 1.

Functional genetic screen for cisplatin-sensitizing siRNAs. A, HNSCC cell line VU-SCC-120 was treated with 18 different concentrations of cisplatin to determinethe IC50 (4.0mmol/L) and IC20 (2.0mmol/L) values.B,Designof high-throughput siRNA screenwithVU-SCC-120 cells.Whitewells contain untransfected control cells,gray wells contain transfection controls (siCONTROL#2 and siPLK1), and black wells contain cells transfected with siRNAs from the genome-wide library.C, Graphic representation of the Z-score distribution. Black dots represent siRNAs from the genome-wide screen. The cut-off for hit list preparation was setat Z��3 (P < 0.001). None of the nontargeting siCONTROL#2 siRNAs (gray dots) reached this threshold.D, Schematic representation of the number of siRNAs thatwere found to significantly sensitize VU-SCC-120 cells to cisplatin, either by Z-score calculation or a Limma-based linear regression model. Comparison of thetwo hit lists showed that 102 siRNAs were identified by both statistical methods.






arrest in G2-M after cisplatin treatment, as is the case after BRCA2knockdown. We compared cell-cycle distributions among cellcultures treated with or without cisplatin (Fig. 3A). We applied500 nmol/L cisplatin, the highest concentration that does notshow an effect on cell survival (Fig. 1A). Indeed, this concentra-tion did not result in cell-cycle arrest. BRCA2 knockdown (nocisplatin added) caused a significant shift toward a G2–Marrest inthe cell-cycle distribution (P < 0.05, two-sided Fisher exact test).This further increased after cisplatin treatment (P < 0.001). For theSHFM1 knockdown cells, we found a small, but not significantshift in cell-cycle distribution in the cultures that were not treatedwith cisplatin (P¼ 0.13, two-sided Fisher exact test in comparisonwith siCONTROL#2-transfected cells), but the proportion of cells

in each phase of the cell cycle was significantly different aftercisplatin incubation (P < 0.02). This indicates that SHFM1 is acritical component in the FA/BRCA pathway, and reduced expres-sion causes increased cisplatin sensitivity and G2–M arrest.

To investigate whether BRCA1, BRCA2, and SHFM1, identifiedas determinants of cisplatin sensitivity by our genome-widesiRNA screen, could be suitable candidates to serve as markersthat predict whether cisplatin-containing chemoradiation is suc-cessful in HNSCC patients, we mined microarray mRNA expres-sion profiles made from 22 normal mucosa/tumor pairs for thesegenes. We found that the expression of BRCA1, BRCA2, andSHFM1was significantly upregulated (P < 0.001) in HNSCC cellscompared with the paired normalmucosa (1.94x� 0.60, 1.27x�

Table 1. List of siRNAs that were confirmed to sensitize VU-SCC-120 (and additional HNSCC cell lines) to cisplatin with at least two separate siRNAs

Gene symbol Origin Confirmed in Accession number Gene name

ALKBH2 Overlap 3 HNSCC cell lines NM_001145374 AlkB, alkylation repair homolog 2 (E. coli)APP Overlap VU-SCC-120 NM_000484 Amyloid beta (A4) precursor protein (Alzheimer disease)ATR Overlap 3 HNSCC cell lines NM_001184 Ataxia telangiectasia and Rad3 relatedBCL11A Overlap VU-SCC-120 NM_138559 B-cell CLL/lymphoma 11A (zinc finger protein)BRCA1 Overlap 3 HNSCC cell lines NM_007294 Breast cancer 1, early onsetBRCA2 Overlap 3 HNSCC cell lines NM_000059 Breast cancer 2, early onsetBRUNOL6 Overlap VU-SCC-120 NM_052840 Bruno-like 6, RNA binding protein (Drosophila)C17ORF35 Overlap VU-SCC-120 NM_003876 Chromosome 17 open reading frame 35CAMK2B Overlap 3 HNSCC cell lines NM_001220 Calcium/calmodulin-dependent protein kinase II betaCDK19 Overlap 3 HNSCC cell lines NM_015076 Cell division cycle 2-like 6 (CDK8-like)CENPT Lin. reg. model VU-SCC-120 NM_025082 Centromere Protein TCHRNA6 Overlap VU-SCC-120 NM_004198 Cholinergic receptor, nicotinic, alpha polypeptide 6CHUK Overlap VU-SCC-120 NM_001278 Conserved helix-loop-helix ubiquitous kinaseCLK3 Overlap 3 HNSCC cell lines NM_001292 CDC-like kinase 3COASY Overlap VU-SCC-120 NM_025233 Coenzyme A synthaseDNAH3 Overlap VU-SCC-120 NM_017539 Dynein, axonemal, heavy polypeptide 3DSCR1 Z-score analysis VU-SCC-120 NM_004414 Down syndrome critical region gene 1ETNK2 Overlap VU-SCC-120 NM_018208 Ethanolamine kinase 2ETV3L Z-score analysis VU-SCC-120 NM_001004341 Ets Variant 3-LikeFANCM Z-score analysis 3 HNSCC cell lines NM_020937 Fanconi anemia, complementation group MFOXL1 Overlap VU-SCC-120 NM_005250 Forkhead box L1ING1L Overlap 3 HNSCC cell lines NM_001564 Inhibitor of growth family, member 2KCNJ2 Overlap 3 HNSCC cell lines NM_000891 Potassium inwardly-rectifying channel, subfamily J, member 2LCE1B Overlap 3 HNSCC cell lines NM_178349 Late cornified envelope 1BLOC115704 Lin. reg. model VU-SCC-120 NM_145245MARK3 Overlap 3 HNSCC cell lines NM_002376 MAP/microtubule affinity-regulating kinase 3MTNR1B Overlap VU-SCC-120 NM_005959 Melatonin receptor 1BMTX1 Overlap 3 HNSCC cell lines NM_002455 Metaxin 1NR1H2 Overlap VU-SCC-120 NM_007121 Nuclear receptor subfamily 1, group H, member 2NUP153 Z-score analysis VU-SCC-120 NM_005124 Nucleoporin 153 kDaOR1D2 Overlap VU-SCC-120 NM_002548 Olfactory receptor, family 1, subfamily D, member 2PALB2 Overlap VU-SCC-120 NM_024675 Partner and localizer of BRCA2PRDM13 Overlap VU-SCC-120 NM_021620 PR domain containing 13PTAFR Overlap 3 HNSCC cell lines NM_000952 Platelet-activating factor receptorPTMA Overlap 3 HNSCC cell lines NM_002823 Prothymosin, alpha (gene sequence 28)RBMS3 Overlap VU-SCC-120 NM_014483 RNA binding motif, single stranded interacting proteinREV3L Overlap 3 HNSCC cell lines NM_002912 REV3-like, catalytic subunit of DNA polymerase zeta (yeast)RHAG Overlap VU-SCC-120 NM_000324 Rhesus blood group-associated glycoproteinRNF26 Overlap VU-SCC-120 NM_032015 Ring finger protein 26SHFM1 Z-score analysis VU-SCC-120 NM_006304 Split hand/foot malformation (ectrodactyly) type 1SLC25A27 Overlap 3 HNSCC cell lines NM_004277 Solute carrier family 25, member 27SLC44A1 Overlap VU-SCC-120 NM_080546 Solute carrier family 44, member 1SSUH2 Lin. reg. model 3 HNSCC cell lines NM_001256748 Ssu-2 Homolog (C. elegans)TADA3L Overlap 3 HNSCC cell lines NM_006354 Transcriptional adaptor 3 (NGG1 homolog, yeast)-likeTF Overlap 3 HNSCC cell lines NM_001063 TransferrinTFF1 Z-score analysis 3 HNSCC cell lines NM_003225 Trefoil factor 1 (breast cancer estrogen-inducible protein)TNFRSF13B Overlap VU-SCC-120 NM_012452 Tumor necrosis factor receptor superfamily, member 13BTNK1 Overlap VU-SCC-120 NM_003985 Tyrosine kinase, nonreceptor, 1TOP2A Lin. reg. model VU-SCC-120 NM_001067 Topoisomerase (DNA) II Alpha 170 kDaTREM2 Overlap VU-SCC-120 NM_018965 Triggering receptor expressed on myeloid cells 2TTBK2 Overlap VU-SCC-120 NM_173500 Tau tubulin kinase 2






0.40, and 1.62x� 0.69 fold, respectively; Fig. 3B). In addition, therange of gene expression varies tremendously among the differenttumor samples, leaving the possibility open that variation in geneexpression might determine the variation in clinical response tocisplatin.

To investigate this possibility, we analyzed the publiclyavailable gene expression dataset of Wichmann and colleagues(64), which contains information on the expression profiles of290 HNSCC tumors and progression-free (PFS) and overallsurvival (OS). All tumors were treated with a range of multi-modal therapies with curative intent. We first identified allprimary tumors that did not contain HPV DNA and/or RNA(n ¼ 121), as this is a major prognostic factor. Next, we selectedall probes targeting the 102 genes that sensitize HNSCC cells forcisplatin as identified by the two different statistical methodsused for analysis of the genome-wide siRNA screen (Fig. 1D).This left us with 127 probes targeting 84 different genes(Supplementary Table S4), and we analyzed whether theexpression of these genes (that sensitize for cisplatin treatment)is significantly associated with patient survival. A (cox) globaltest (65) indicated that our gene set shows a significant corre-

lation with the overall survival of the patients (P ¼ 0.039).Next, we fitted a multivariable cox model with ridge regular-ization using package glmnet (66) and R version 3.1.3. Theaccuracy of this model was assessed with a leave one out crossvalidation. Across the first 3 years after treatment, this gave anintegrated area under the curve (iAUC; ref. 67) of 0.60. Indi-vidual HRs resulting from the cross-validation where divided inthree equal groups (Supplementary Table S5) and the survivalof each of these groups was visualized by Kaplan–Meier anal-ysis (Fig. 4). Together, these results indicate that the 84-gene setidentified in our cisplatin-sensitivity screen is correlated withoverall survival and confirmed the importance of our in vitrofindings in an independent study in a patient cohort.

DiscussionResistance to platinum-containing chemoradiation may be a

limiting factor in the successful treatment of advanced stageHNSCC. It would be of significant clinical benefit to under-stand the underlying factors that cause response, because thismay reveal biomarkers for response prediction and pave the

Figure 2.

Functional relationship of the cisplatin-sensitizing siRNA targets. A, Comprehensive pathway analysis of all genes present in our genome-wide siRNA libraryand the associated Z-scores by IPA (Ingenuity Pathway Analysis) showed thatmany siRNAs that target members of the FA/BRCA pathway significantly enhance thecisplatin response in VU-SCC-120. A darker gray color is associated with a lower Z-score. B, String analysis of siRNAs that were confirmed to increasecisplatin sensitivity in all three HNSCC cell lines (n¼ 21) showed that the only cluster that could be identified, harbors DNA repair genes. C, The siRNA SMARTpoolstargeting BRCA1 (left), BRCA2 (middle), and SHFM1 (right) were deconvoluted in VU-SCC-120 cells. Dark gray bars represent IC0-treated cells (no cisplatin),lighter bars correspond to IC20-treated cells. Bars represent cell viability as compared with the IC0 value of each transfection; error bars, SD. � , P < 0.05 comparedwith siCONTROL#2 IC20 value of the same gene (Student t test).






way to a more personalized treatment. In this study, we iden-tified determinants of response to cisplatin in an unbiased loss-of-function screen.

In a previous publication, we described that DNA-boundplatinum is the most important determinant of cisplatin sen-sitivity in HNSCC cell lines (58), and not intracellular platinumlevels. These data already pointed in the direction that theactivity of the DNA cross-link repair machinery might be thecritical factor in HNSCC that determines cisplatin response. Inthis study, we identified the genes involved in repair of DNAcrosslinks, such as BRCA1, BRCA2, and ATR, as most importantgenes that modulate cisplatin response. Patients suffering fromFA/BRCA pathway deficiency are predisposed for squamous cellcancer most particularly in the head and neck likely by the toxiceffect of aldehydes as natural cross-linkers (68). The samepathway apparently determines response to DNA crosslinkersused as cytotoxic drugs. Examination of the activity level of theFA/BRCA pathway prior to cisplatin-containing treatment isan option to predict upfront whether addition of cisplatin to

therapy would succeed, for example by screening for geneexpression or mutations. However, mutations in these particu-lar genes are uncommon in HNSCC (COSMIC database; cancer.sanger.ac.uk/cancergenome/projects/cosmic). This highlightsthe necessity for the identification of biomarkers that mightdisclose whether the FA/BRCA pathway and/or the involvedgenes are at high activity or not. Such biomarkers could be theexpression of the genes themselves as the first analysis of theexpression data suggests. A functional DNA crosslinking assay orimaging of radioactive cisplatin in tumor DNA might be thebiomarker of choice to personalize treatment of patients thatwill not benefit from cisplatin-containing chemoradiation pro-tocols. Alternatively, the combination of chemoradiation andan agent that inhibits DNA crosslink repair might enhanceefficacy of cisplatin and lead to higher chemoradiation responserates, although also toxicity will likely increase.

We identified SHFM1 as an important player in the responseto cisplatin treatment. SHFM1 is known to bind to BRCA2(61, 62) and to stabilize this protein (60), although this could

AsiCONTROL#2 BRCA2 SHFM1

0 nmol/L

G1: 79.2%

S: 10.1%

G2−M: 9.9%

G1: 66.1%

S: 12.8%

G2−M: 19.9%

G1: 66.3%

S: 15.9%

G2−M: 16.3%

G1: 56.4%

S: 16.2%

G2−M: 25.9%

G1: 47.8%

S: 20.9%

G2−M: 28.9%

G1: 75.1%

S: 11.0%

G2−M: 12.9%500 nmol/L

0%

25%

50%

75%

100%

siCON 0 nmol/L siCON 500 nmol/L BRCA2 0 nmol/L BRCA2 500 nmol/L SHFM1 0 nmol/L SHFM1 500 nmol/L

G2−MSG1

***

*

B0.8

0.6

0.4

0.2

BRCA1

Mucosa Tumor

Rel

ativ

e ex

pres

sion

Mucosa Tumor

BRCA21.4

1.2

1.0

0.8

0.6

0.4

Rel

ativ

e ex

pres

sion

Mucosa Tumor

SHFM14.0

3.0

2.0

1.0

0.0

Rel

ativ

e ex

pres

sion

*

*

*

Figure 3.

Functional relationship of the cisplatin-sensitizing siRNA targets. A, Cell-cycle distribution of VU-SCC-120 cells transfected with siCONTROL#2, BRCA2, orSHFM1-targeting siRNA SMARTpools. Cultures were not treated with cisplatin (0 nmol/L) or treated with 500 nmol/L cisplatin during 72 hours. Percentages of cellsin designated cell-cycle phases are given and are schematically represented in the bottom panel (� , P < 0.02; �� , P < 0.001). Error bars, SD. B, Microarraygene expression data from 22 patient-derived pairs of tumor tissue and healthy mucosa were examined for BRCA1 (top), BRCA2 (middle), and SHFM1 (bottom)expression. Thick lines within the boxplots represent the median values. The expression of all three genes was significantly higher in the tumor samplesthan in the normal mucosa (� raw P < 0.001, moderated t test implemented in Limma).






not be confirmed in every study (63). SHFM1 has been previ-ously found to sensitize HeLa cells to cisplatin (35). In thatstudy, the authors also focused on members of the FA/BRCApathway, which indicates that this pathway is not only impor-tant in the cisplatin response in SCCs from the head and neck,but also in other squamous cell carcinomas. In addition, arecent study in ovarian cancer showed that SHFM1 has a role incisplatin resistance (69) and that SHFM1 levels are elevated intumor cells as compared with the nontumorigenic neighboringcells. This observation serves as a platform to use SHFM1expression as a biomarker for ovarian cancer. Also in breastcancer, high SHFM1 expression was found to correlate to poorprognosis and poor relapse-free survival (70). We found thatthe SHFM1 expression levels are overall elevated in HNSCC ascompared with healthy mucosa (Fig. 3B), and that the level ofexpression varies tremendously.

We noticed that several genes that are known to influence thecisplatin response were not found in our genome-wide siRNAscreen. For some of these genes, it might have occurred that thesiRNA-induced gene knockdown was insufficient, as we deter-mined for ATM (Supplementary Fig. S3A). However, the knock-down of CTR1 was significant (Supplementary Fig. S3B), but wecould not find clues for the involvement of this gene in cisplatinresponse (Supplementary Fig. S3C), as we could not before (58).This is surprising asmany studies implicated an important role forCTR1 in cisplatin uptake (71–73). Because these studies were notperformed in HNSCC, one might argue that other genes thanCTR1 are of greater importance in cisplatin response in HNSCC.This may also be the reason why we identified some of the SLCfamily members that are previously implicated in cisplatin(SLC25A27 and SLC44A1; Table 1; Supplementary Table S2),but not all, such as SLC7A11, which proved to be important incisplatin resistance in bladder cancer (74) and collecting duct

carcinoma of the kidney (75). Furthermore, it might haveoccurred that compensatory mechanisms abolishe the effect ofknockdown expression of a particular gene (76, 77). For example,the functions of AKT1, AKT2, and AKT3 seem partly redundant,implying that knockdown of either one of these genes might notresult in a phenotype as the other two genes are able to take oversome of the functions of the silenced gene.

Finally, we identified novel genes that may influence cisplatinresponse in HNSCC cells, and that have not been describedbefore. An example is KCNJ2, a potassium inwardly-rectifyingchannel, that has a small but significant effect on cisplatin sen-sitivity and we were able confirm this in all three cell lines used(Supplementary Fig. S4A). Also the blocking of the channelingfunction of KCNJ2, by barium chloride (78–80), showed that cellcultures are significantly sensitized to cisplatin (SupplementaryFig. S4B). However, the observed effect is small and not compa-rable with that of knockdown of, for example, BRCA2.

We investigated the importance of the genes we identified ascisplatin response determinants in an independent, publiclyavailable expression dataset based on a cohort of HNSCCpatientstreated with multimodality treatment (64). We found that theexpression of genes in our gene set correlates with patient survival(integrated AUC ¼ 0.60) and that higher expression of the genesresult in worse patient survival. This indicates that the gene set weidentified in vitro to influence cisplatin sensitivity also contributesin vivo to HNSCC treatment outcome. Larger studies of morehomogenous patient series will be required to validate thesefindings, and to establish that some of these genes might beexploited as biomarker.

Taken together, our data show that the application of anunbiased genome-wide functional genetic screen leads to theidentification of unexpected and novel determinants of cisplatinresponse, which might be exploited as biomarkers for clinicalprediction of chemoradiation response. Our data will redirectclinical research towards reliable response predictors for plati-num-containing compounds applied in chemoradiation andsystemic therapy.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: S.R. Martens-de Kemp, B.J.M. Braakhuis,R.H. BrakenhoffDevelopment of methodology: S.R. Martens-de Kemp, I.H. van der Meulen,V.W. van Beusechem, R.H. BrakenhoffAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): S.R. Martens-de Kemp, A. Brink, I.H. van der Meulen,V.W. van BeusechemAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S.R. Martens-de Kemp, A. Brink, R.X. de Menezes,D.E. te Beest, C.R. Leemans, B.J.M. Braakhuis, R.H. BrakenhoffWriting, review, and/or revision of the manuscript: S.R. Martens-de Kemp,I.H. van der Meulen, C.R. Leemans, V.W. van Beusechem, B.J.M. Braakhuis,R.H. BrakenhoffAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): S.R. Martens-de Kemp, I.H. van der Meulen,R.H. BrakenhoffStudy supervision: B.J.M. Braakhuis, R.H. Brakenhoff

AcknowledgmentsThe authors would like to thank Dirk J. Kuik for assistance during the

statistical analysis of the genome-wide siRNA screen. This study was performed

Figure 4.

Patient survival is correlated with expression of cisplatin-sensitizing genes.The Kaplan–Meier curves are based on the cross-validated HRs calculated withthe multivariable cox model using the expression of 84 genes that wereidentified in our genome-wide siRNA screen. The patient cohort was divided inthree equal groups based on these HRs. Survival curves indicate that patientswith higher (cross-validated) HR have a worse overall survival.






within the framework of CTMM, the Centre for Translational Molecular Med-icine AIRFORCE project. High-throughput screens were conducted at the RNAInterference Functional Oncogenomics Laboratory (RIFOL) core facility at theVUmc Cancer Center Amsterdam.

Grant SupportR.H. Brakenhoff received a grant from CTMM, the Centre for Translational

Molecular Medicine (AIRFORCE project, grant 03O-103). V.W. van Beuse-chem received financial support from the Stichting VUmc CCA for establish-

ing the RNA Interference Functional Oncogenomics Laboratory (RIFOL) corefacility.

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 July 18, 2016; revised November 18, 2016; accepted December 6,2016; published OnlineFirst December 15, 2016.

References1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of

worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer2010;127:2893–917.

2. Leemans CR, Braakhuis BJ, Brakenhoff RH. The molecular biology of headand neck cancer. Nat Rev Cancer 2011;11:9–22.

3. Pignon JP, le MA, Maillard E, Bourhis J. Meta-analysis of chemotherapy inhead and neck cancer (MACH-NC): an update on 93 randomised trials and17,346 patients. Radiother Oncol 2009;92:4–14.

4. Chen SW, Yang SN, Liang JA, Lin FJ, Tsai MH. Prognostic impact of tumorvolume in patientswith stage III-IVA hypopharyngeal cancer without bulkylymph nodes treated with definitive concurrent chemoradiotherapy. HeadNeck 2009;31:709–16.

5. Doweck I, Denys D, Robbins KT. Tumor volume predicts outcome foradvanced head and neck cancer treated with targeted chemoradiotherapy.Laryngoscope 2002;112:1742–9.

6. Hoebers FJ, Pameijer FA, de BJ, Heemsbergen W, Balm AJ, Schornagel JH,et al. Prognostic value of primary tumor volume after concurrent chemor-adiation with daily low-dose cisplatin for advanced-stage head and neckcarcinoma. Head Neck 2008;30:1216–23.

7. Knegjens JL, Hauptmann M, Pameijer FA, Balm AJ, Hoebers FJ, de Bois JA,et al. Tumor volume as prognostic factor in chemoradiation for advancedhead and neck cancer. Head Neck 2010;33:375–82.

8. Kurek R, Kalogera-Fountzila A, Muskalla K, Dafni U, Schnabel T, Kober B,et al. Usefulness of tumor volumetry as a prognostic factor of survival inhead and neck cancer. Strahlenther Onkol 2003;179:292–7.

9. Studer G, Lutolf UM, El-Bassiouni M, Rousson V, Glanzmann C. Volu-metric staging (VS) is superior to TNM and AJCC staging in predictingoutcome of head and neck cancer treated with IMRT. Acta Oncol2007;46:386–94.

10. Tsou YA, Hua JH, Lin MH, Tsai MH. Analysis of prognostic factors ofchemoradiation therapy for advanced hypopharyngeal cancer–does tumorvolume correlate with central necrosis and tumor pathology? ORLJ Otorhinolaryngol Relat Spec 2006;68:206–12.

11. van den BroekGB, Rasch CR, Pameijer FA, Peter E, van den BrekelMW, TanIB, et al. Pretreatment probability model for predicting outcome afterintraarterial chemoradiation for advanced head and neck carcinoma.Cancer 2004;101:1809–17.

12. de Jong MC, Pramana J, Knegjens JL, Balm AJ, van den Brekel MW,Hauptmann M, et al. HPV and high-risk gene expression profiles predictresponse to chemoradiotherapy in head and neck cancer, independent ofclinical factors. Radiother Oncol 2010;95:365–70.

13. Handra-Luca A, Hernandez J, Mountzios G, Taranchon E, Lacau-St-Guily J,Soria JC, et al. Excision repair cross complementation group 1 immuno-histochemical expression predicts objective response and cancer-specificsurvival in patients treated by cisplatin-based induction chemotherapy forlocally advanced head and neck squamous cell carcinoma. Clin Cancer Res2007;13:3855–9.

14. Homma A, Furuta Y, Oridate N, Nakano Y, Kohashi G, Yagi K, et al.Prognostic significance of clinical parameters and biological markers inpatients with squamous cell carcinoma of the head and neck treated withconcurrent chemoradiotherapy. Clin Cancer Res 1999;5:801–6.

15. Jun HJ, Ahn MJ, Kim HS, Yi SY, Han J, Lee SK, et al. ERCC1 expression as apredictivemarker of squamous cell carcinoma of the head and neck treatedwith cisplatin-based concurrent chemoradiation. Br J Cancer 2008;99:167–72.

16. Koukourakis MI, Giatromanolaki A, Kakolyris S, Sivridis E, Georgoulias V,Funtzilas G, et al. Nuclear expression of human apurinic/apyrimidinicendonuclease (HAP1/Ref-1) in head-and-neck cancer is associated with

resistance to chemoradiotherapy and poor outcome. Int J Radiat OncolBiol Phys 2001;50:27–36.

17. Lavertu P, Adelstein DJ, Myles J, Secic M. P53 and Ki-67 as outcomepredictors for advanced squamous cell cancers of the head and neck treatedwith chemoradiotherapy. Laryngoscope 2001;111:1878–92.

18. Osman I, Sherman E, Singh B, Venkatraman E, Zelefsky M, Bosl G, et al.Alteration of p53 pathway in squamous cell carcinoma of the head andneck: impact on treatment outcome in patients treated with larynx pres-ervation intent. J Clin Oncol 2002;20:2980–7.

19. van den Broek GB, Wildeman M, Rasch CR, Armstrong N, Schuuring E,Begg AC, et al. Molecular markers predict outcome in squamous cellcarcinoma of the head and neck after concomitant cisplatin-based che-moradiation. Int J Cancer 2009;124:2643–50.

20. Warnakulasuriya S, Jia C, Johnson N, Houghton J. p53 and P-glyco-protein expression are significant prognostic markers in advanced headand neck cancer treated with chemo/radiotherapy. J Pathol 2000;191:33–8.

21. Akervall J,GuoX,QianCN, Schoumans J, Leeser B, Kort E, et al. Genetic andexpression profiles of squamous cell carcinoma of the head and neckcorrelate with cisplatin sensitivity and resistance in cell lines and patients.Clin Cancer Res 2004;10:8204–13.

22. Dumur CI, Ladd AC, Wright HV, Penberthy LT, Wilkinson DS, Powers CN,et al. Genes involved in radiation therapy response in head and neckcancers. Laryngoscope 2009;119:91–101.

23. DuongC,Greenawalt DM, Kowalczyk A, CiavarellaML, Raskutti G,MurrayWK, et al. Pretreatment gene expression profiles can be used to predictresponse to neoadjuvant chemoradiotherapy in esophageal cancer. AnnSurg Oncol 2007;14:3602–9.

24. Ganly I, Talbot S, Carlson D, Viale A, Maghami E, Osman I, et al. Iden-tification of angiogenesis/metastases genes predicting chemoradiotherapyresponse in patients with laryngopharyngeal carcinoma. J Clin Oncol2007;25:1369–76.

25. Hanna E, Shrieve DC, Ratanatharathorn V, Xia X, Breau R, Suen J, et al. Anovel alternative approach for prediction of radiation response of squa-mous cell carcinoma of head and neck. Cancer Res 2001;61:2376–80.

26. Maher SG, GillhamCM,Duggan SP, Smyth PC,Miller N,MuldoonC, et al.Gene expression analysis of diagnostic biopsies predicts pathologicalresponse to neoadjuvant chemoradiotherapy of esophageal cancer. AnnSurg 2009;250:729–37.

27. Pramana J, vandenBrekelMW, vanVelthuysenML,Wessels LF,NuytenDS,Hofland I, et al. Gene expression profiling to predict outcome afterchemoradiation in head and neck cancer. Int J Radiat Oncol Biol Phys2007;69:1544–52.

28. Schauer M, Janssen KP, Rimkus C, Raggi M, Feith M, Friess H, et al.Microarray-based response prediction in esophageal adenocarcinoma.Clin Cancer Res 2010;16:330–7.

29. Bao S, Wu Q, McLendon RE, Hao Y, Shi Q, Hjelmeland AB, et al. Gliomastem cells promote radioresistance by preferential activation of the DNAdamage response. Nature 2006;444:756–60.

30. Berns K,Hijmans EM,Mullenders J, Brummelkamp TR, Velds A, HeimerikxM, et al. A large-scale RNAi screen in human cells identifies new compo-nents of the p53 pathway. Nature 2004;428:431–7.

31. Gobeil S, Zhu X, Doillon CJ, Green MR. A genome-wide shRNA screenidentifies GAS1 as a novel melanoma metastasis suppressor gene. GenesDev 2008;22:2932–40.

32. Jian R, Cheng X, Jiang J, Deng S, Hu F, Zhang J. A cDNA-based randomRNAinterference library for functional genetic screens in embryonic stem cells.Stem Cells 2007;25:1904–12.






33. Krishnan MN, Ng A, Sukumaran B, Gilfoy FD, Uchil PD, Sultana H, et al.RNA interference screen for human genes associated with West Nile virusinfection. Nature 2008;455:242–5.

34. Arora S, Bisanz KM, Peralta LA, Basu GD, Choudhary A, Tibes R, et al. RNAiscreening of the kinome identifies modulators of cisplatin response inovarian cancer cells. Gynecol Oncol 2010;118:220–7.

35. Bartz SR, Zhang Z, Burchard J, ImakuraM,MartinM, Palmieri A, et al. Smallinterfering RNA screens reveal enhanced cisplatin cytotoxicity in tumorcells having both BRCA network and TP53 disruptions. Mol Cell Biol2006;26:9377–86.

36. Bauer JA, Ye F, Marshall CB, Lehmann BD, Pendleton CS, Shyr Y, et al.RNA interference (RNAi) screening approach identifies agents thatenhance paclitaxel activity in breast cancer cells. Breast Cancer Res2010;12:R41–56.

37. Chen S, Blank JL, Peters T, Liu XJ, Rappoli DM, Pickard MD, et al. Genome-wide siRNA screen for modulators of cell death induced by proteasomeinhibitor bortezomib. Cancer Res 2010;70:4318–26.

38. Harradine KA, Kassner M, Chow D, Aziz M, Von Hoff DD, Baker JB, et al.Functional genomics reveals diverse cellular processes that modulatetumor cell response to oxaliplatin. Mol Cancer Res 2011;9:173–82.

39. Iorns E, Turner NC, Elliott R, Syed N, Garrone O, Gasco M, et al. Identi-fication of CDK10 as an important determinant of resistance to endocrinetherapy for breast cancer. Cancer Cell 2008;13:91–104.

40. Mackeigan JP, Murphy LO, Blenis J. Sensitized RNAi screen of humankinases and phosphatases identifies new regulators of apoptosis andchemoresistance. Nat Cell Biol 2005;7:591–600.

41. Swanton C, Marani M, Pardo O, Warne PH, Kelly G, Sahai E, et al.Regulators of mitotic arrest and ceramide metabolism are determinantsof sensitivity to paclitaxel and other chemotherapeutic drugs. Cancer Cell2007;11:498–512.

42. Turner NC, Lord CJ, Iorns E, Brough R, Swift S, Elliott R, et al. A syntheticlethal siRNA screen identifying genes mediating sensitivity to a PARPinhibitor. EMBO J 2008;27:1368–77.

43. Whitehurst AW, Bodemann BO, Cardenas J, Ferguson D, Girard L, PeytonM, et al. Synthetic lethal screen identification of chemosensitizer loci incancer cells. Nature 2007;446:815–9.

44. PosthumaDe Boer J, van EgmondPW,HelderMN, deMenezes RX, Cleton-Jansen AM, Belien JA, et al. Targeting JNK-interacting-protein-1 (JIP1)sensitises osteosarcoma to doxorubicin. Oncotarget 2012;3:1169–81.

45. Pruitt KD, Tatusova T, Maglott DR. NCBI Reference Sequence (RefSeq): acurated non-redundant sequence database of genomes, transcripts andproteins. Nucleic Acids Res 2005;33:D501–4.

46. Lin CJ, Grandis JR, Carey TE, Gollin SM, Whiteside TL, Koch WM, et al.Head and neck squamous cell carcinoma cell lines: establishedmodels andrationale for selection. Head Neck 2007;29:163–88.

47. HermsenMA, Joenje H, Arwert F,WeltersMJ, Braakhuis BJ, BagnayM, et al.Centromeric breakage as amajor cause of cytogenetic abnormalities in oralsquamous cell carcinoma. Genes Chromosomes Cancer 1996;15:1–9.

48. Brenner JC, GrahamMP, Kumar B, Saunders LM, Kupfer R, Lyons RH, et al.Genotyping of 73 UM-SCC head and neck squamous cell carcinoma celllines. Head Neck 2010;32:417–26.

49. Martens-de Kemp SR, Nagel R, Stigter-van Walsum M, van der Meulen I,van Beusechem V, Braakhuis BJ, et al. Functional genetic screens identifygenes essential for tumor cell survival in head and neck and lung cancer.Clin Cancer Res 2013;19:1994–2003.

50. BoutrosM, Bras LP,HuberW. Analysis of cell-based RNAi screens. GenomeBiol 2006;7:R66.

51. Smyth GK. Linear models and empirical Bayes methods for assessingdifferential expression in microarray experiments. Stat Appl Genet MolBiol 2004;3:Article3.

52. Smyth GK. Limma: linear models for microarray data. In: Gentleman R,Carey VJ, Dudoit S, Irizarry RA, Huber W, editors. Bioinformatics andcomputational biology solutions using r and bioconductor. New York, NY:Springer; 2005. p.397–420.

53. Benjamini Y, Hochberg Y. Controlling the false discovery rate - a practicaland powerful approach to multiple testing. J Roy Stat Soc B Met1995;57:289–300.

54. deMenezes RX, Boer JM, vanHouwelingenHC.Microarray data analysis: ahierarchical T-test to handle heteroscedasticity. Appl Bioinformatics 2004;3:229–35.

55. Aerts JL, Gonzales MI, Topalian SL. Selection of appropriate control genesto assess expression of tumor antigens using real-time RT-PCR. Biotechni-ques 2004;36:84–1.

56. Livak KJ, Schmittgen TD. Analysis of relative gene expression data usingreal-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods2001;25:402–8.

57. Martens-de Kemp SR, Brink A, Stigter-van Walsum M, Damen JM, Rusten-burg F, Wu T, et al. CD98 marks a subpopulation of head and necksquamous cell carcinoma cells with stem cell properties. Stem Cell Res2013;10:477–88.

58. Martens-deKempSR,DalmSU,Wijnolts FM,BrinkA,HoneywellRJ,PetersGJ,et al.DNA-boundplatinum is themajor determinant of cisplatin sensitivity inhead and neck squamous carcinoma cells. PLoS One 2013;8:e61555.

59. Spankuch-Schmitt B, Bereiter-Hahn J, KaufmannM, Strebhardt K. Effect ofRNA silencing of polo-like kinase-1 (PLK1) on apoptosis and spindleformation in human cancer cells. J Natl Cancer Inst 2002;94:1863–77.

60. Li J, Zou C, Bai Y, Wazer DE, Band V, Gao Q. DSS1 is required for thestability of BRCA2. Oncogene 2006;25:1186–94.

61. MarstonNJ, RichardsWJ,HughesD, BertwistleD,Marshall CJ, AshworthA.Interaction between the product of the breast cancer susceptibility geneBRCA2 and DSS1, a protein functionally conserved from yeast to mam-mals. Mol Cell Biol 1999;19:4633–42.

62. Yang H, Jeffrey PD, Miller J, Kinnucan E, Sun Y, Thoma NH, et al. BRCA2function in DNA binding and recombination from a BRCA2-DSS1-ssDNAstructure. Science 2002;297:1837–48.

63. Gudmundsdottir K, Lord CJ, Witt E, Tutt AN, Ashworth A. DSS1 is requiredfor RAD51 focus formation and genomic stability in mammalian cells.EMBO Rep 2004;5:989–93.

64. Wichmann G, Rosolowski M, Krohn K, Kreuz M, Boehm A, Reiche A, et al.The role of HPV RNA transcription, immune response-related gene expres-sion and disruptive TP53mutations in diagnostic and prognostic profilingof head and neck cancer. Int J Cancer 2015;137:2846–57.

65. Goeman JJ, Oosting J, Cleton-Jansen AM, Anninga JK, van HouwelingenHC. Testing association of a pathway with survival using gene expressiondata. Bioinformatics 2005;21:1950–7.

66. SimonN, Friedman J, Hastie T, Tibshirani R. Regularization paths for Cox'sproportional hazards model via coordinate descent. J Stat Softw2011;39:1–13.

67. Heagerty PJ, Zheng Y. Survival model predictive accuracy and ROC curves.Biometrics 2005;61:92–105.

68. Langevin F,CrossanGP,Rosado IV,ArendsMJ, Patel KJ. Fancd2 counteractsthe toxic effects of naturally produced aldehydes in mice. Nature 2011;475:53–8.

69. MaYY, LinH,Chang FM,Chang TC, Trieu T, PridgenHI, et al. Identificationof the deleted in split hand/split foot 1 protein as a novel biomarker forhuman cervical cancer. Carcinogenesis 2013;34:68–78.

70. Rezano A, Kuwahara K, Yamamoto-Ibusuki M, KitabatakeM,Moolthiya P,Phimsen S, et al. Breast cancers with high DSS1 expression that potentiallymaintains BRCA2 stability have poor prognosis in the relapse-free survival.BMC Cancer 2013;13:562.

71. Holzer AK, Samimi G, Katano K, NaerdemannW, Lin X, Safaei R, et al. Thecopper influx transporter human copper transport protein 1 regulates theuptake of cisplatin in human ovarian carcinoma cells. Mol Pharmacol2004;66:817–23.

72. Ishida S, Lee J, Thiele DJ, Herskowitz I. Uptake of the anticancer drugcisplatin mediated by the copper transporter Ctr1 in yeast and mammals.Proc Natl Acad Sci U S A 2002;99:14298–302.

73. Lin X, Okuda T, Holzer A, Howell SB. The copper transporter CTR1regulates cisplatin uptake in Saccharomycescerevisiae. Mol Pharmacol2002;62:1154–9.

74. Drayton RM, Dudziec E, Peter S, Bertz S, Hartmann A, Bryant HE, et al.Reduced expression of miRNA27a modulates cisplatin resistance in blad-der cancer by targeting the cysteine/glutamate exchanger SLC7A11. ClinCancer Res 2014;20:1990–2000.

75. Wang J, Papanicolau-Sengos A, Chintala S, Wei L, Liu B, Hu Q, et al.Collecting duct carcinomaof the kidney is associatedwithCDKN2Adeletionand SLC family gene up-regulation. Oncotarget 2016;7:29901–15.

76. Degtyarev M, De MA, Orr C, Lin J, Lee BB, Tien JY, et al. Akt inhibitionpromotes autophagy and sensitizes PTEN-null tumors to lysosomotropicagents. J Cell Biol 2008;183:101–16.






77. Gill C, Dowling C, O'Neill AJ, Watson RW. Effects of cIAP-1, cIAP-2 andXIAP triple knockdown on prostate cancer cell susceptibility to apoptosis,cell survival and proliferation. Mol Cancer 2009;8:39.

78. Alagem N, Dvir M, Reuveny E. Mechanism of Ba(2þ) block of a mouseinwardly rectifying Kþ channel: differential contribution by two discreteresidues. J Physiol 2001;534:381–93.

79. Chun TU, EpsteinMR, DickM, Andelfinger G, Ballester L, Vanoye CG, et al.Polymorphic ventricular tachycardia and KCNJ2mutations. Heart Rhythm2004;1:235–41.

80. Tsuboi M, Antzelevitch C. Cellular basis for electrocardiographic andarrhythmic manifestations of Andersen-Tawil syndrome (LQT7). HeartRhythm 2006;3:328–35.






2017;16:540-550. Published OnlineFirst December 15, 2016.Mol Cancer Ther Sanne R. Martens-de Kemp, Arjen Brink, Ida H. van der Meulen, et al. Response in Head and Neck Cancer by Functional GenomicsThe FA/BRCA Pathway Identified as the Major Predictor of Cisplatin

Updated version

10.1158/1535-7163.MCT-16-0457doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2016/12/14/1535-7163.MCT-16-0457.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/16/3/540.full#ref-list-1

This article cites 79 articles, 21 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/16/3/540.full#related-urls

This article has been cited by 1 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://mct.aacrjournals.org/content/16/3/540To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-16-0457

http://mct.aacrjournals.org/content/suppl/2016/12/14/1535-7163.MCT-16-0457.DC1

http://mct.aacrjournals.org/content/16/3/540.full#ref-list-1

http://mct.aacrjournals.org/content/16/3/540.full#related-urls

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

mailto:[email protected]

http://mct.aacrjournals.org/content/16/3/540


the fa/brca pathway identiﬁed as the major predictor of ......companion diagnostics and cancer...

Documents