Malignant pleural mesothelioma: genome-wide expression patterns reflecting general resistance mechanisms and a proposal of novel targets.

Abstract

Malignant pleural mesothelioma is an asbestos-related multi-resistant tumour with increasing incidence worldwide. Well-characterized snap-frozen normal parietal, visceral pleura and mesothelioma samples were analysed with Affymetrix Human Genome U133 Plus 2.0 GeneChip oligoarray of 38500 genes. We discovered a close relation between gene profile and resistance towards topoisomerase poisons, alkylating agents, antitubulines, antifolates, platinum compounds and radiation therapy. Target genes of chemo- (e.g. TOP2A, BIRC5/Survivin and proteasome) and radiotherapy (e.g. BRCA2, FANCA, FANCD2, CCNB1 and RAD50) were significantly overexpressed. The Fanconi anemia/BRCA2 pathway, responsible for homologous recombination DNA repair appears as a key pathway in both chemo- and radio-resistance of mesothelioma. Leukocyte trans-endothelial migration gene down-regulation could partly explain resistance against immunological therapies. Gene expression features found in other resistant cancer types related to DNA repair and replication are shared by mesothelioma and could represent general features of tumour resistance. Targeted suppression of some of those key genes and pathways combined with chemotherapy or radiation could improve the outcome of mesothelioma therapy. We propose CHEK1, RAD21, FANCD2 and RAN as new co-targets for mesothelioma treatment. The pro-angiogenic AGGF1 mRNA and protein was highly overexpressed in all tumours and may serve as a target for anti-angiogenic treatment. Overexpression of NQO1 may render mesothelioma sensitive to the novel compound beta-Lapachone.

Full text

Page 1

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
Lung Cancer xxx (2009) xxx–xxx
Contents lists available at ScienceDirect
Lung Cancer
journal homepage: www.elsevier.com/locate/lungcan
Malignant pleural mesothelioma: Genome-wide expression patterns reflecting
general resistance mechanisms and a proposal of novel targets
Oluf Dimitri Røea,b,∗, Endre Anderssenb, Helmut Sandeckc, Tone Christensenb,
Erik Larssond, Steinar Lundgrena,b
aDepartment of Oncology, St. Olavs Hospital, Trondheim, Norway
bDepartment of Cancer Research and Molecular Medicine (IKM), Norwegian University of Science and Technology (NTNU), Trondheim, Norway
cDepartment of Pathology and Medical Genetics, St. Olavs Hospital, Trondheim, Norway
dDepartment of Laboratory Medicine, Children’s and Women’s Health (LBK), Norwegian University of Science and Technology (NTNU), Trondheim, Norway
a r t i c l ei n f o
Article history:
Received 8 October 2008
Received in revised form 12 January 2009
Accepted 17 March 2009
Keywords:
Chemo-resistance
DNA repair
Drug target
Fanconi anemia
Homologous recombination
Microarray
Radiation resistance
a b s t r a c t
Malignant pleural mesothelioma is an asbestos-related multi-resistant tumour with increasing incidence
worldwide. Well-characterized snap-frozen normal parietal, visceral pleura and mesothelioma samples
were analysed with Affymetrix Human Genome U133 Plus 2.0 GeneChip oligoarray of 38500 genes.
We discovered a close relation between gene profile and resistance towards topoisomerase poisons,
alkylating agents, antitubulines, antifolates, platinum compounds and radiation therapy. Target genes
of chemo- (e.g. TOP2A, BIRC5/Survivin and proteasome) and radiotherapy (e.g. BRCA2, FANCA, FANCD2,
CCNB1 and RAD50) were significantly overexpressed. The Fanconi anemia/BRCA2 pathway, responsible
forhomologousrecombinationDNArepairappearsasakeypathwayinbothchemo-andradio-resistance
ofmesothelioma.Leukocytetrans-endothelialmigrationgenedown-regulationcouldpartlyexplainresis-
tance against immunological therapies. Gene expression features found in other resistant cancer types
related to DNA repair and replication are shared by mesothelioma and could represent general features
of tumour resistance. Targeted suppression of some of those key genes and pathways combined with
chemotherapy or radiation could improve the outcome of mesothelioma therapy. We propose CHEK1,
RAD21, FANCD2 and RAN as new co-targets for mesothelioma treatment. The pro-angiogenic AGGF1
mRNA and protein was highly overexpressed in all tumours and may serve as a target for anti-angiogenic
treatment. Overexpression of NQO1 may render mesothelioma sensitive to the novel compound beta-
Lapachone.
© 2009 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
Median survival of malignant mesothelioma is currently 12
months [1], an aggressive tumour derived from cells of the
pleura, peritoneum or tunica vaginalis testis, where pleural loca-
tion accounts for about 70% of the cases [2]. Epithelial subtype
is an important positive prognostic factor in contrast to the sar-
comatous and mixed subtypes. It is among the most chemo- and
radio-resistant malignant tumours [3] and no curative treatment
is available, but occasionally good responders and long-term sur-
vivorsareseen.Pemetrexed,amulti-folateinhibitorcombinedwith
cisplatin is the only treatment that has shown increased median
survival in a Phase III study, from 9.3 to 12.1 months [4]. Stud-
ies unraveling the biology of mesothelioma and identifying novel
molecular targets for treatment are thus urgently warranted. We
∗Corresponding author. Tel.: +47 73900650.
E-mail address: oluf.roe@stolav.no (O.D. Røe).
recently did genome-wide expression analysis of human pleural
mesothelioma versus normal parietal pleura samples that portraits
the aggressive and resistant biology of mesothelioma (Røe et al., in
preparation). Here we focus on genes and pathways with signifi-
cant differential expression from the same material, that could be
related to the drug and irradiation resistance of pleural mesothe-
lioma and propose some future targets.
2. Materials and methods
2.1. Mesothelioma and control patients
Patients diagnosed between 2003 and 2005 were included after
obtaining informed written consent. Diagnostic biopsies and mate-
rial for gene expression were taken from adjacent locations with
needles by Computer Tomography and/or ultrasound guidance.
Diagnostic samples were formalin-fixed and paraffin-embedded.
Material for gene expression analysis was snap-frozen in liquid
nitrogen within two minutes. Biopsies of morphologically normal
0169-5002/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.lungcan.2009.03.016

Page 2

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
2
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
pleurawereobtainedfrompersonswhounderwentVideo-Assisted
Thoracoscopy (VATS) for recurrent pneumothorax, after obtain-
ing patient history and informed consent. Parietal pleura that was
stripped from the thoracic wall and visceral pleura dissected from
the wedge-resections of the lung, were snap-frozen in liquid nitro-
gen within two minutes. Mesothelioma diagnosis was carried out
byseniorpathologistsandre-examinedbyH.Sandeck,byincluding
a standard panel of antibodies for immunohistochemistry as well
as by using supplementary antibodies.
The study protocol and biobank were approved by the Regional
Ethics Committee, the National Health Department and the
National Data Inspectory.
2.2. RNA-extraction and microarray experiments
Methods used for RNA extraction were optimized to assure a
highqualityRNAfromthesmallneedlebiopsiesofthetumours.The
finaltechniquechosenwashomogenizationoffrozentissue2×50s,
withMagnaLyser(RocheDiagnostics)followingthemanufacturer’s
procedure, using 700?L lysis buffer (Roche Diagnostics, Germany)
as it gave higher RNA yield. The material was then incubated for
30minatroomtemperatureandcentrifugedat13000×Gfor2min.
350?L of the supernatant was used for further RNA isolation. Man-
ual isolation with High Pure RNA Tissue Kit (Roche Diagnostics,
Germany) according to the producer’s protocol was performed.
Quality control of RNA was done with NanoDrop (Saveen & Werner
AB, Sweden) and Bioanalyzer (Agilent Technologies, Inc., USA).
Gene expression analysis was performed by the Affymetrix
GeneChip system according to the manufacturer’s Eukaryote Two-
Cycle protocol, starting with 75ng deep frozen total RNA. Labelled
cRNA was hybridized to the Affymetrix Human Genome U133
Plus 2.0 GeneChip (Affymetrix, Santa Clara, CA, USA), of 38500
genes and 47000 trancripts, allowing genome-wide expression on
a single array. The GeneChips were scanned using the GeneChip
Scanner 3000 (Affymetrix). Quality controls were assessed using
the GCOS v1.4 software, according to the manufacturer’s manual
(Affymetrix).AllexperimentshavebeensubmittedtoArrayExpress
registered with accession number E-MTAB-47.
2.3. Microarray statistical analysis
The raw probe set intensities were normalised by robust multi
array average (RMA). Quality control was done using density plots,
box plots and principal component analysis (PCA). Differentially
expressed genes were detected using a Bayesian linear model
method [5]. P-values were corrected for multiple testing using the
method of Benjamini and Hochberg [6,7] and genes with corrected
P-values <0.05 were taken as significant. The lists of significant
genes were tested for over-representation in KEGG (Kyoto Ency-
clopedia of Genes and Genomes) PATHWAYS [8], and GO (gene
ontology) [9] terms using Fishers exact test. The distribution of the
gene expression pattern in significant pathways was visualised in
the loading space of a bridge-partial least squares regression (PLS)
model [10].
2.4. Validation
Cell specific expressions of proteins encoded by six selected
genes, were validated by immunohistochemistry. The following
antibodies were tested on fixed tissues adjacent to samples sub-
jected to microarray (gene symbols in parentheses). Thymidylate
synthase (TYMS) (Millipore, USA) dilution 1:50, VG5Q (AGGF1)
(Abcam, Cambridge UK) dilution 1:500, Chk1 (CHEK1) (Epito-
mics, California, USA), dilution 1:10, overnight incubation at −4◦C,
NQO1 (NQO1) (Zymed Laboratories, Carlsbad, CA, USA) dilution
1:50, RAD21 (RAD21) (Abcam, Cambridge, UK) dilution 1:500 and
Mesothelin (MSLN) (Novocastra Laboratories, Newcastle, UK) was
analysedwithdilution1:10,overnightincubationat−4◦C.Selected
positive and negative controls were included for all antibodies.
3. Results
3.1. General expression patterns
Mesothelioma, parietal and visceral pleural samples showed
distinct expression profiles (Fig. 1) and the following results are
derived from profiling analysis of mesothelioma versus only pari-
etal pleura samples. Parietal pleura was used as the main reference
because mesothelioma primarily develops from the parietal layer
[11].
3.2. Functional gene expression
GO analysis showed significant overexpression of genes in
mesothelioma reflecting several important biological functions,
among them regulation of progression through cell cycle, DNA
replication and repair as well as microtubule cytoskeleton orga-
nization and biogenesis (Table 1). Within the DNA repair entity,
genes related to double-strand break repair were over-represented
(Table 1). Several key genes encoding proteins are known as
treatment targets, but they also confer to chemo-resistance as
TOP2A (topoisomerase 2A), TYMS (thymidylate synthase), BIRC5
(survivin), NQO1 (NAD(P)H dehydrogenase, quinone 1), and radio-
resistance by several DNA repair and damage checkpoint genes
among others BRCA2 (breast cancer susceptibility gene 2), CHEK1
(checkpointkinase1),FANCA(FanconianemiaA),FANCD2(Fanconi
anemia D2), RAD21 (resistant to radiation 21), RAD50 (resistant to
radiation 50) and several proteasome genes were overexpressed in
tumour (Table 1). A novel pro-angiogenic gene AGGF1 (angiogenic
factor with G patch and FHA domains 1) was also significantly over-
expressed. Several genes encoding detoxifying enzymes, among
them the multidrug resistance gene ABCB1 (ATP-binding cassette
sub-family B member 1) were down-regulated, as were the leuko-
Fig. 1. Partial least square analysis of the total gene expression of numbered
mesothelioma (T), parietal pleura (PP) and visceral pleura (PV) tissue samples. The
three tissues show different expression profiles. Patient 8 had the most similar
expression profile to PP. This patient has partial remission and is a long-term sur-
vivor.Patient10withsamplestakenfromtwodifferentsitesoftumourhadthemost
aberrantexpressionprofile.Thispatientalsohadthemostrapidgrowthofherradio-
and chemo-resistant mesothelioma.

Page 3

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
3
Table 1
Overexpressed gene ontology (GO) entities with corresponding overexpressed genes (Genes Up) in mesothelioma versus parietal pleura. Genes on Chip is the number of
genes from each entity represented on the gene chip. These are genes involved in regulation of the cell cycle, replication, repair, tRNA synthesis, proteasome and microtubuli
synthesis and function, where not only possible mechanisms of resistance but also putative treatment targets may be hypothesized. Selected genes are discussed in the text.
GO term Genes UpGenes on ChipCorrected P-valueGene symbols
GO:0000074 regulation of progression
through cell cycle
50 504 2.29E−09PHB, RAN, CCT7, PSMD2, PSMD1, PCNA, NME2, NME1, MYBL2,
CDC123, CDK2AP1, CCT2, BIRC5, SRPK1, CCNB2, CDC20, TIMELESS,
RB1, BUB1B, CDC6, ZWINT, CDC7, CDC25A, CHEK1, MATK, CENPF,
BRCA2, CADM1, UBE2V2, AATF, BUB1, CCNO, CDC2, NUDC, CDK7,
MKI67, PRKCA, NUSAP1, PTMS, MCTS1, GTPBP4, GMNN, HYPE,
TCF19, CDC23, PPAPDC1B, BCCIP, E2F2, CCNB1, ESPL1
PHB, PCNA, TOP2A, RRM1, MCM3, MCM6, CDK2AP1, MCM2, TYMS,
SSBP1, MSH6, RNASEH2A, RFC5, CDC6, RFC4, RBM14, FEN1, GINS1,
GLI2, DNA2L, PRIM2, PTMS, GTPBP4, GMNN, ORC6L, GINS2, MCM4
SETMAR, RAD21, PCNA, TOP2A, CIB1, SHFM1, TYMS, MSH6, RFC5,
RFC4, RAD51AP1, RBM14, RAD54L, FEN1, XRCC4, CHEK1, BRCA2,
RAD50, UBE2V2CCNO, CDK7, FANCI, FANCF, RAD18, GTF2H2,
NSMCE1, UHRF1, BCCIP, EME1, FANCA, FANCD2
RAD21, CIB1, SHFM1, RAD51AP1, FEN1, XRCC4, BRCA2, RAD50,
UBE2V2
AARS, MARS, XPOT, XPO5, NSUN2, TRNT1
RAN, BUB1B, ZWINT, NDC80, KIF23, AURKA, KIF2C, BBS4, NUSAP1,
DYNC1H1, CLASP1, ESPL1
KIF5B, KIF23, KIF2C, KIFC3, DYNC1I2, BBS4, DYNLRB1, KIF18A,
DYNC1H1, KIF14
PSMB2, PSMD2, PSMC2, PSMD1, PSMC3, SHFM1, PSMA4, PSMC1,
PSMD11, PSMC5, PSME3, PSMD14, PSMA7
GO:0006260 DNA replication27181 3.51E−08
GO:0006281 DNA repair 312373.38E−08
GO:0006302 double-strand break repair9 280.0001
GO:0000049 tRNA binding
GO:0000226 microtubule cytoskeleton
organization and biogenesis
GO:0003777 microtubule motor activity
6 17
69
0.0033
0.0005 12
10 780.0165
GO:0000502 proteasome complex 13 462.14E−06
Fig. 2. Protein expression of three overexpressed genes with possible therapeutic implications. (A, C and E) Normal parietal pleura. (B, D and F) Mesothelioma with epithelial
and sarcomatous component. (A and B) CHEK1, negative normal pleura. All tumours except one had >15% tumour nuclei positivity. (C and D) RAD21, the normal mesothelial
cells nuclei are weakly stained, some endothelial cells and histiocytes were positive. Both the epithelial and sarcomatous tumour components were intensely positive. (E and
F) NQO1, normal mesothelium stained intensely but also budding capillaries (arrow), stroma generally negative. Intense expression in tumour of the epithelial component,
few sarcomatous cells were positive.

Page 4

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
4
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
cyte trans-endothelial migration pathway and signal transduction
genes (Fig. 5).
3.3. Verification by protein expression
All samples adjacent to the samples subjected to microarray,
except control no. 3 where there was no sufficient material, were
analysedbyimmunohistochemistryforproteinexpressionofgenes.
Five significantly overexpressed samples were tested, of which
three are shown here (Fig. 2).
Thymidylate synthasewas
brane/cytoplasma and some nuclei of most tumour cells whereas
parietal pleural samples were completely negative.
RAD21 antibody demonstrated nuclear staining in 40 to >80% of
displayed tumour cells, except but one negative case. Some areas of
normalmesotheliumnucleiwereweaklystainedandsomestromal
cells were positive. In one control 60% of displayed cells, predomi-
nantly immune cells, were positive.
Chk1 antibody stained >15% of epithelial mesothelioma nuclei
except the sample from the only long-term survivor that had <1%
stained nuclei. No nuclear staining was seen in the control samples.
expressedin the mem-
NQO1 antibody stained nucleus and/or cytoplasm of >75% of all
tumourcells,fewsarcomatouscellswerepositive.Normalendothe-
lial cells were positive, especially in areas of neovascularization in
inflammatory areas of one case and one control.
VG5Q (AGGF1) stained nucleus and/or cytoplasm in >75% of all
tumour cells, the sarcomatous component included (not shown).
Strong positive nuclear expression of VG5Q was seen in normal
mesothelial and endothelial cells as well as in histiocytes.
MSLN (mesothelin) was not differentially expressed, but is
known to be highly expressed in mesotheliomas and is an impor-
tantfuturetreatmenttarget[12].Wefoundthatalltumoursamples
had high membrane expression of mesothelin of variable intensity
(not shown), but mesothelial cell membrane and submesothelial
fibroblasts were also strongly positive, underscoring one reason
why MSLN gene was not overexpressed in tumour.
4. Discussion
Our recent genome-wide profiling study of malignant pleural
mesothelioma versus normal parietal pleura showed new expres-
sion patterns that translate into highly relevant biological features
Table 2
Review of response and survival of mesothelioma by chemotherapy and irradiation; their known targets related to significantly overexpressed genes and gene ontology (GO)
entities or pathways discovered in the present study. Treatments that achieved more than 10 months of median survival are noted in italics (*HR, homologous recombination
genes; NHEJ, non-homologous end joining genes, see text,**Response was measured as disease stabilization>6 months).
Single agent Phase I–IIRespiration
rate (%)
Median survival
(months)
Ref.Molecular and
functional targets
Overexpressed
genes, P<0.05
Overexpressed GO entities,
genes (n), P<0.05
Camptotecins Top I,
anti-replication,
DNA breaks
CHEK1,
proteasome, HR,
NHEJ*
DNA-replication (27) DNA
repair (31)
Irinothecan, Topotecan
Taxanes/vinca-alcaloids
0
≤10Baas [3]
Anti-tubuline, Top
I, anti-replication
BUB1B, survivin,
RAN, KIF, PTK2
DNA-replication (27)
microtubule cytoskeleton
organization and
biogenesis (12)
microtubule motor activity
(10)
Paclitaxel
Vinblastine/-desine/-cristine
Vinorelbine
Alkylating agents
Cyclo-phosphamide
Aziridinyl benzoquinone
Anthracyclines
0
0–3
20
≤10
≤10
10.6
Baas [3]
Baas [3]
Baas [3]
NQO1, DNA breaks NQO1, HR, NHEJ*
0
0
Sorensen et al. (2005)
Baas [3]
≤10
Top IIa, DNA breaks TOP2A, survivin,
HR, NHEJ*
DNA-replication (27), DNA
repair (31)
Doxorubicin
Liposomal doxorubicin
Mitomycin
Detorubicin
Mitoxathrone
Platinum analogues
11
0–20
21
43
0–13
≤10
13
≤10
17
≤10
Baas [3]
Baas [3]
Baas [3]
Colbert et al. (1985)
Eisenhauer et al. (1986)
DNA cross-
linking/breaks
Survivin, HSP90B1,
PRKCI, laminin, HR,
NHEJ*
DNA repair (31)
Cisplatin
Carboplatin
Anti-metabolites
4–31
5.4–34
≤10
≤10
Mintzer et al. (1985)
Raghavan et al. (1990)
TYMS, GART, DHFR,
RRM1, DNA
damage
TYMS, RRM1 HR,
NHEJ*
DNA-replication (27), DNA
repair (31)
Metothrexate (high dose)
5-FU
Gemcitabine
Pemetrexed
Targeted-Erotinib
Targeted-Ranpirnase
37
5
7
14
0
4.9
11
≤10
≤10
10.7
10
11.3
Solheim et al. (1992)
Baas [3]
Baas [3]
Scagliotti et al. (2003)
Garland et al. (2007)
Mikulski et al. (2002)
EGFR
destroys tRNA
Survivin
FARSA, GARS, DARS2
DNA-replication (27)
Aminoacyl tRNA
biosynthesis (6)
Anti-apoptosis
Thalidomide27.5**
7.2Baas et al. [70] TNF, bFGF, NFkBNFkB-stimulating
genes
Survivin, CCNB1
HR, NHEJ*
TYMS, HR, NHEJ*
Radiation0.3
≤10Ung et al. (2006)Top I, EGFR,
survivin, HR, NHEJ*
TYMS, GART, DHFR,
DNA
cross-linking/breaks
DNA-replication (27), DNA
repair (31)
DNA-replication (27) DNA
repair (31)
Phase III Cisplatin/pemetrexed 41 12.1Vogelzang et al. [4]

Page 5

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
5
as cell cycle, apoptosis, DNA repair and circadian rhythms (Table 1,
Røeetal.inpreparation).Herewewillfocusongenesandpathways
involved in anti-tumour resistance mechanisms and their implica-
tions for treatment.
Mesothelioma has been treated with several singlet, doublet
and triplet drug combinations [3] as well as with ionizing radia-
tion [13] with very low response rates and survival (Table 2) over
the last three decades. When tumour has intrinsic resistance to
such a wide range of therapies, it may imply activity of multiple
resistanceandtumoursurvivalmechanisms.Identificationofthose
could give insight into mesothelioma resistance and hopefully also
give some insight into tumour resistance in general. If the mRNA
expression patterns detected here reflect the respective protein
expressions, as our validation has indicated, and based on knowl-
edge of molecular mechanisms of action and resistance towards
anti-cancer treatment, we identified significant overexpression of
both target and resistance related genes of the chemotherapeuti-
cal classes that have been tried in mesothelioma. These include
taxanes, vinca alcaloids, camptotecins, anthracyclines, alkylating
agents, antimetabolites and platinum compounds, the novel tar-
getedproteasomeinhibitorsandranpirnase,aswellasradiotherapy
(Table 2), and are discussed below.
4.1. Antitubulines—the taxanes and vinca alkaloids
Monotherapy with taxanes or vinca alkaloids have with the
exception of vinorelbine minimal effect on mesothelioma in ear-
lier trials (Table 2). Vinorelbine combined with cisplatin has shown
a promising effect in a recent Phase II study [14]. The main
target of these drugs is stabilizing the dynamic of the micro-
tubule system in the interphase and M-phase, inducing DNA
damage, chromosomal imbalance and subsequently apoptosis [15].
In our material, several genes related to M-phase and microtubule
cytoskeleton organization and biogenesis were significantly over-
expressed. Genes involved in the spindle assembly checkpoint that
ensures attachment of the spindle microtubules of all sister chro-
matides were overexpressed, as the BUB1 (budding uninhibited 1)
and BUB1B. Overexpression of BUB1B significantly correlated with
higher histological grade, advanced pathological stage, and high
cell proliferation in bladder cancer and predicted tumour recur-
rence and disease progression [16]. BUB1 is a possible negative
prognostic factor in mesothelioma [17]. Survivin, encoded by BIRC5
(baculoviralIAPrepeat-containing5)has,besidesitsanti-apoptotic
function,aroleinmicrotubuledynamicsandcontrolbipolarspindle
formation [18]. Its overexpression is associated to paclitaxel resis-
tance, and inhibition of survivin increased paclitaxel induced cell
death. Importantly, survivin also seems to control Ran, encoded
by RAN (member RAS oncogene family) that was overexpressed.
Ran is a small GTPase regulator of mitotic spindle formation and
importantly it is overexpressed in human cancer as compared with
normal tissues. Gene silencing of RAN induces aberrant mitotic
spindle formation, mitochondrial dysfunction, and apoptosis [19].
Loss of Ran in normal cells is well tolerated and does not result
in mitotic defects or loss of cell viability, making it a suitable
anti-cancer target. DYNC1H1 (dynein, cytoplasmic 1, heavy chain
1) was overexpressed, encoding dynein, a microtubule-activated
ATPase that serve to convert chemical energy into mechanical
energy involved in intracellular motility, protein sorting between
apical and basolateral surfaces, and redistribution of organelles
like endosomes and lysosomes [20]. It is also highly expressed
in gliomas and has been related to glioma migration and prolif-
eration [21]. BBS4 (Bardet-Biedl syndrome 4), also overexpressed,
localizes to the centriolar satellites of centrosomes and basal bod-
ies of primary cilia. Silencing of BBS4 induces deanchoring of
centrosomal microtubules, arrest in cell division and apoptotic
cell death [22]. The genes encoding centromere proteins (CENPA,
CENPF, CENPI, CENPN, CEP170) as well as kinesins (KIF14, KIF18A,
KIF23, KIF5B, KIF2C, KIF3C) were overexpressed. Kinesin inhibi-
tion induced mitotic cell death in docetaxel resistant and sensitive
ovariancancercelllines[23],indicatingthatkinesinesnotonlycon-
tribute to resistance but also act as a potential target that could
increase taxane sensitivity. NDC80 (NDC80 homolog, kinetochore
complex component), encoding Hec1 (highly expressed in cancer)
wasoverexpressed.Hec1isoneofseveralproteinsinvolvedinspin-
dle checkpoint signalling, detecting unaligned chromosomes and
causingprometaphasearrestuntiltheproperbipolarattachmentof
chromosomes is achieved. Hec1 protein depletion induced mitotic
catastrophe in cervix adenocarcinoma and glioblastoma cell lines
[24], and is regarded as a possible gene therapy target. Cyclin B1,
an important cell cycle protein encoded by CCNB1 is concentrated
at the centrosomes and spindle microtubules, interacting with
the molecular chaperone Hsp90 (heat shock protein 90) and the
microtubule-associated protein mini spindles CKAP5 (cytoskele-
ton associated protein 5) [25]. Both CCNB1 and HSP90B1 (heat
shockprotein90kDabeta(Grp94),member1)wereoverexpressed,
the last was recently proposed as a target for mesothelioma [26].
PTK2 (protein tyrosin kinase 2) that was overexpressed, encodes
the focal adhesion kinase that besides being important in micro-
tubuleorganizationhasantiapoptoticandpro-angiogenicfunctions
[27]. Currently small molecules inhibitors of focal adhesion kinase
are tested on various cancer models showing tumour inhibition
and human trials are ongoing [28]. Furthermore, when PTK2 was
silenced docetaxel-mediated apoptosis was augmented in ovarian
cancer cells [29]. Indirectly, overexpression of PTK2 could thus con-
fer to taxane resistance, consequently also in mesothelioma.
4.2. Antimetabolites
Antimetabolites as methotrexate and pemetrexed have con-
ferred more tumour responses than most other agents in
mesothelioma (Table 2). Thymidylate synthase encoded by TYMS
is a key protein that catalyzes the methylation of deoxyuridylate
(dUMP)todeoxythymidylate(dTMP)thatmaintainsthedTMPpool
critical for DNA replication and repair. This enzyme has also been
recognised as an important target of the multifolate pemetrexed,
but its overexpression is also a resistance factor of antimetabolites
[30]. Low expression of thymidylate synthase of freshly explanted
human tumour specimens correlated to high in vitro chemosensi-
tivity to pemetrexed [31], in line with low TYMS and thymidylate
synthase expression found in our patient with long-term survival
and continuous effect of pemetrexed and high expression in the
most aggressive case (Fig. 6). In patients with resected stage I non-
smallcelllungcancerthymidylatesynthaseproteinoverexpression
was a positive predictive factor [32], so this issue is clearly not yet
fully elucidated. Moreover, TYMS inhibitors increase the levels of
dUTP (deoxyuridine triphosphate) which in turn incorporated in
theDNA,whichistoxic.ThisismainlycounteractedbyUNG2(uracil
DNA-glycosylase) encoded by the CCNO/UNG2 (cyclin O) gene, that
eliminate uracil from the DNA by cleaving the N-glycosylic bond
and initiate the base-excision repair (BER) pathway [33]. This con-
fersastrandbreakthatmaynotberepairedduetothelackofdTMP,
inducing cell death [34], but it is still not clear whether this repair
mechanism also could confer to cell survival. In our material the
CCNO was overexpressed, and our long-term survivor with excel-
lent effect of pemetrexed had CCNO mRNA expression 40% higher
than the patient with the most aggressive disease, indicating that
BER could play a role in pemetrexed induced cell death (Fig. 3).
The novel antimetabolite gemcitabine targets RRM1 (ribonu-
cleotide reductase subunit M1) and this compound showed very
lowsingle-drugeffectonmesothelioma(Table2).Inlungadenocar-
cinoma RRM1 mRNA overexpression was recently associated with
significantlylowerresponseratesandevensurvival[35].RRM1was

Page 6

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
6
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
Fig. 3. Schematic presentation based on our data of genes overexpressed in mesothelioma related to DNA repair (green). Damage by chemotherapy or radiation induce two
known lethal lesions of the DNA, interstrand cross-links (ICL) and double-strand breaks (DSB). Both can be repaired by homologous recombination (HR) and the Fanconi
anemia/BRCA2 pathway, where most of the overexpressed DNA repair genes were clustered. The error-prone non-homologous end joining (NHEJ) can repair single and
double-strand breaks without a template, represented by four overexpressed genes. MSH6, a mismatch repair (MMR) gene was also overexpressed and two genes from the
“short-patch” and “long-patch” base-excision repair (BER/SP, BER/LP) pathway were represented as well as a nucleotide excision repair (NER) gene. CHEK1 was overexpressed
which is activated by DNA damage, inducing cell cycle arrest allowing the repair process to proceed and escape cell death. Disruption of HR, NHEJ and CHEK1 by various
compounds, sensitize tumours to cross-linking agents and radiation (putative targets with bold letters). Proteasome function is crucial for normal function of the Fanconi
anemia/BRCA2pathwayandthusproteasomeoverexpressionislinkedtoDNArepair.RAD50hasaroleintelomeremaintenanceandmaybeveryimportantformesothelioma,
as no other genes with similar function were overexpressed.
overexpressed in mesothelioma, a novel finding that may relate to
its gemcitabine resistance.
4.3. Topoisomerase I poisons—camptotecines
Camptotecines target the cleavable complex between the topoi-
somerase I, and the DNA inducing irreversible double-strand
breaks. Irinotecan and topotecan have no anti-tumour effect on
mesothelioma (Table 2). Several mechanisms of resistance have
been proposed [36], two of those coincide with our findings; over-
expression of CHEK1 encoding the G2/M DNA repair checkpoint
Chk1 protein, and proteasome overexpression. Loss of Chk1 func-
tionandproteasomeinhibitionenhancedtoxicityofcamptotecines
[37,38], rendering these possible co-targets for mesothelioma.
4.4. Topoisomerase II poisons—anthracyclines
Anthracyclines are topoisomerase II poisons that stabilizes the
cleavable complexes of DNA inducing double-strand breaks and
forms covalent adducts inducing similar damage as the cross-
linking of platinum compounds [39]. TOP2A (topoisomerase 2A)
protein expression is a marker of invasive breast cancer aggressive-
ness [40], but the relation between gene amplification and protein
expression is inconsistent [41]. TOP2A gene amplification in HER-
2 amplified early breast cancer was a positive predictive marker
for anthracyclin treatment, but surprisingly, TOP2A deletions also
predicted higher anthracyclin sensitivity [42]. TOP2A was over-
expressed in our material, and although anthracyclines have had
low response rates, there has been some clinical benefit of pegy-
lated doxorubicin [43] and detorubicin (no longer in clinical use) as
monotherapy. This last compound had surprisingly two complete
remissions and a median survival of 17 months in mesothelioma
(Table 2) and 36% response rate in metastatic melanoma [44].
Combined therapy of pegylated doxorubicin, gemcitabine and car-
boplatin in a large Nordic Phase II study showed increased survival
[45], indicating that combined targeting may be more efficient.
TOP2A overexpression has not been described in mesothelioma
previously and its expression should be explored in relation to
anthracycline sensitivity. As to DNA repair see Section 4.7.
4.5. Cross-linkers and alkylating agents—platinum analogues and
azaridinyl benzoquinone
Platinum analogues have some activity in mesothelioma, but
havenotimprovedsurvivalasmonotherapy(Table2).Severalother
systems and genes involved in platinum resistance coincide with
our findings including extracellular matrix factors (laminin), apop-
tosis inhibitors (survivin), chaperones (HSP90B1), cell signalling
(PRKCI) and cell cycle related factors [46]. Down-regulation of
CAV1 conferring to resistance was also detected. Azaridinyl ben-
zoquinone is a prodrug that is reduced by NQO1 and alkylates DNA
conferring strand breaks. It also has no clinical effect on mesothe-
lioma, even if NQO1 is overexpressed in our material and thus
theoretically should work more efficiently [47]. Most important
thoughseemstobetheDNArepairmechanismdescribedinSection
4.7.

Page 7

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311; No.of Pages12
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
7
4.6. Ionizing radiation
Ionizing radiation has several effects on cells, but the main
anti-tumour effects are believed to be single- or double-strand
DNA breaks directly, or indirectly by releasing free oxygen radi-
cals. Radio-resistance, early recurrence and metastasis are related
to high CCNB1 expression in head and neck cancer [48]. We found
thisaninherentfeatureofmesothelioma,andcyclinB1overexpres-
sion has been shown in aquired radio-resistance, possibly through
the activation of NFkB and several antiapoptotic mechanisms [49].
Survivin overexpression has also been shown to increase radio-
resistance [50]. The role of DNA repair is discussed below.
4.7. DNA repair and resistance against cross-linkers,
topoisomerase inhibitors and ionizing radiation
The DNA is the main cytotoxic target of cisplatin and carbo-
platinbyinductionofsingleanddouble-strandDNAbreaksthrough
adducts and cross-linking, leading to cell death through apoptosis
[51]. To counteract a platinum compound cytotoxic attack requires
ahighlycomplexrepaircascadeofmanymechanisms.Recently,the
Fanconi anemia/BRCA2 (FA) pathway has been attributed a role as
a coordinator of this cascade [52].
Homologous recombination (HR) is dependent on a complex of
several FANC proteins that monoubiquitinate the FANCD2 which
then interacts with BRCA2 and further with the members of the
RAD family, BCCIP and SHFM1 (Table 1 and Fig. 3). Several of these
key genes were overexpressed and none were down-regulated.
People with FANC or BRCA2 mutations have increased risk for
malignancy, but when they develop tumours that are hypersensi-
tivetocross-linkingagents[53].Inhibitionoftheproteinkinasethat
monoubiquitinatestheFANCD2bythenaturalcompoundcurcumin
sensitizes ovarian cancer cells to cisplatin [54].
Non-homologous end joining (NHEJ) is an error-prone repair
of double-strand breaks where a key gene XRCC4 was overex-
pressed. Natural vanillin suppresses NHEJ by a similar kinase and
also sensitizes ovarian cancer cells for cisplatin [55]. This also indi-
cates that inhibition of the two repair mechanisms HR and NHEJ
could sensitize mesothelioma to platinum compounds. Defect mis-
match repair (MMR) confers to platinum resistance, but only one
of the MMR genes, MSH6 was overexpressed here. The nucleotide
excisionrepair(NER)whenoverexpressedconferstocisplatinresis-
tance but only one gene of this mechanism was overexpressed
(GTF2H2/p44).
HR and NER seem to be the most important mechanisms con-
ferring anthracycline resistance, similar to cisplatin resistance.
HR deficient cancer cells (V-C8) were 3–6-fold more sensitive to
doxorubicin and 6–12-fold more sensitive to cisplatin than their
parental proficient cells (V79) [39]. Interestingly, this first cell type
was deficient in BRCA2 that was overexpressed in our mesothe-
liomas, indicating that BRCA2 suppression also could increase
anthracycline effect in mesothelioma.
DNA repair has been shown to play an important role in
irradiation sensitivity and resistance, and inherited gene defects
in humans with loss of repair functions confer hypersensitivity
to radiotherapy, as the Fanconi anemia and BRCA2 gene muta-
tions. There are sporadic clinical reports on this but has been
shown experimentally [56,57]. Moreover, the RAD genes are the
mammalian counterparts of yeast genes that confer radiation
hypersensitivity when mutated or silenced. The exact mechanisms
of action in humans are not known, but the RAD genes over-
expressed here play important roles in HR, NHEJ and even, as
RAD50, in telomerase maintenance [58]. Proteasome inhibition
has also been shown to increase radiosensitivity, and recently it
was shown that proteasome function is required for repair foci
formation, thus linking proteasome directly to HR repair (Fig. 3).
One of the proteasome genes overexpressed here, PSMD14, when
depleted, severely decreased proteasome proteolytic activity and
led to significant inhibition of FANCD2 monoubiquitination [59].
UHRF1 (ubiquitin-like protein containing PHD and RING domain 1)
is a novel human radio-susceptibility gene with unknown mode
of action but its suppression clearly sensitizes human cell lines
to radiation [60]. Overexpression of these genes in mesothelioma
could be very important factors in both its radio-resistance as well
as its platinum resistance. Furthermore, our findings are consistent
witharecentstudyofmetastaticversusnon-metastaticmelanomas
being profiled genome-wide at an early stage [61]. Overexpres-
sion of DNA-repair genes were highly significant in the melanomas
that did metastasize within four years and the main repair path-
ways represented were the Fanconi anemia pathway and HR, as
found in the present study. We identified 13 out of their 44 over-
expressed genes (CHEK1, EME1, FANCA, FANCD2, GTF2H2, MSH6,
PCNA, RAD18, RAD54L, RAD51AP1, RFC4, RFC5, TOP2A) and one
out of their three down-regulated genes CRY2 (cryptochrome 2),
a negative regulator of the circadian rhythm (Røe et al., in prepara-
tion). Similar results were found in lymph node negative aggressive
breast cancer [62] and bladder carcinoma [63]. Finally, p53 sup-
pression by deletion or mutation, or inactivation by SV40 LargeT
Antigen, HPV E6 protein, has been shown to boost HR gene expres-
sioninhumantumourcells>100fold[64].P53expressionisusually
normal in mesothelioma, but CDKN2AIP, the function of which
is activating the p53 [65] was significantly down-regulated (not
shown). If p53 inactivation leads to HR overexpression, this could
be an important mechanism both for genetic instability and resis-
tance. Taken together, the Fanconi anemia/BRCA2 seems to be a key
pathway for both radio- and chemo-resistance, and thus should be
explored as a target and co-target for radio- and chemo-sensitizing
[66].
4.8. Targeted therapies
Theubiquitine–proteasomepathwayisoverexpressedinseveral
tumours and confers tumour resistance [67]. Microarray anal-
ysis of human biphasic versus epithelial mesothelioma of the
peritoneum showed that the more aggressive biphasic type had
overexpressed proteasome genes [68]. Bortezomib, a proteasome
inhibitor approved for second-line treatment of multiple myeloma,
had anti-tumour effect on mesothelioma cells in vitro [68] and in
a tumour xenograft study on mice [69]. We found 13 proteasome
genes overexpressed and none down-regulated, supporting a role
of proteasome inhibitors in mesothelioma. Combination treatment
withbortezomibwithcisplatininaPhaseIIstudyisalreadyplanned
[69]. DNA damage response and Fanconi anemia repair activation
is connected to proteasome function, and proteasome suppres-
sion inhibits DNA-repair (Fig. 3) [59]. Combined proteasome and
DNA repair inhibition could thus theoretically increase the effect of
chemotherapy and radiation.
Ranpirnase, a ribonuclase of the Northern Leopard frog that
acts through degradation of tRNA leading to apoptosis, showed a
slight activity in mesothelioma (Table 2). We found genes encoding
enzymes of DNA and RNA recycling overexpressed (salvage path-
way) as well as genes that load amino acids on tRNA (Fig. 4). EGFR
suppressorgefitinibhasbeentriedbutshowedonly4%responsesin
mesotheliomaandnoincreasedsurvivalasmonotherapyinspiteof
97% histological expression. Erlotinib treatment had similar results
by monotherapy. EGFR mRNA was not significantly overexpressed
in our study.
Thalidomide was shown to induce stable disease for more
than 6 months in 27.5% of a mesothelioma population where 50%
were pretreated, with a median survival of 7 months [70], indi-
cating a tumour stabilizing effect. It is an anti-angiogenic and
immunonomodulatorycompoundinhibitingtheTNFalpha(tumour

Page 8

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311; No.of Pages12
8
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
Fig. 4. Aminoacyl-tRNA biosynthesis genes overexpressed in mesothelioma versus normal parietal pleura. Red are overexpressed genes, the light green are not differentially
expressed (P<0.05, in KEGG PATHWAYS, Kanehisa et al. [8]). tRNA is the target of ranpirnase and overexpresson of these could play a role in resistance towards this com-
pound(MTFMT,mitochondrialmethionyl-tRNAformyltransferase[EC:2.1.2.9];AARS,alanyl-tRNAsynthetase[EC:6.1.1.7];FARSA,phenylalanyl-tRNAsynthetase,alphasubunit
[EC:6.1.1.20]; GARS, glycyl-tRNA synthetase [EC:6.1.1.14]; MARS, methionyl-tRNA synthetase [EC:6.1.1.10]; DARS2, aspartyl-tRNA synthetase 2, mitochondrial [EC:6.1.1.12]).
necrosis factor alpha), FGF2 (basic fibrobast growth factor), NFkap-
paB and stimulating cytotoxic T-cells [71]. One can speculate that
the effect in mesothelioma could be due to the inhibition of NFkap-
paB stimulating genes, overexpressed in our material (Table 2), and
astheTNFgenewasdown-regulatedandtheFGF2notdifferentially
expressed (not shown). Stimulation of cytotoxic T-lymphocytes
could also be a complementary factor, but this could be obstructed
by down-regulated leukocyte trans-endothelial migration genes
(see Section 4.9).
4.9. Leukocyte trans-endothelial migration
Another intriguing finding was that genes supporting trans-
endothelialmigrationof leukocytes
among them were PECAM1, ESAM and JAM2. PECAM1/CD31
(platelet/endothelial cell adhesion molecule) (Fig. 5) is a member
of the immunoglobulin superfamily that is expressed on the sur-
face of platelets, monocytes, neutrophiles, T-cell subsets and is also
a major constituent of the endothelial cell intercellular junction.
It modulates multiple functions besides trans-endothelial migra-
tion, integrin-mediated cell adhesion, angiogenesis, apoptosis,
cell migration and negative regulation of immune cell signalling
[72]. ESAM (endothelial cell adhesion molecule) is specifically
expressed at endothelial tight junctions and on platelets, and
its down-regulation decreases neutrophil extravasation [73].
JAM2/JAMB (junctional adhesion molecule) inhibition decreases
leukocyte infiltration [74]. Dysfunction due to down-regulation of
these genes could hinder the entrance of immunological cells in
tumour resulting in a barrier to immunologic tumour rejection and
thus to immunological therapies. This coincides with results from
an immunotherapy trial showing good responses in mesothelioma,
but only in small tumours [75]. Unfortunately very few patients
are diagnosed with early stage disease.
were down-regulated,
4.10. Detoxifying enzymes
NQO1isadetoxifyingreductase,wherelackoffunctionisrelated
to increased benzene oncogenesis. It is overexpressed in many
tumours [76] as well as in our material. This is a putative target as
described below. MGST2 (microsomal glutathione s-transferase),
members of the glutathione s-transferase family GSTM1, GSTM2
and GSTM5, Cytochrome P450 family members CYP39A1 and
CYP46A1 and PTGS2/COX2, all with detoxifying activities, were
down-regulated, indicating a minor role of these mechanisms in
theintrinsicresistanceofmesothelioma[77,78].NotablytheABCB1,
coding for the multidrug resistance gene 1 was down-regulated
and thus may not play an important role in treatment resistance
of mesothelioma.
4.11. Some putative treatment targets
Chk1 is a key protein controlling the G2/M checkpoint and DNA
repair as well as playing a role in radiation- and chemo-resistance
(Fig. 3) [79]. Cisplatin resistant hepatoma in mice was cured in 80%
oftheanimalsbycombinationofCHEK1silencingandlowdosecis-
platin[80],showingthepotentialofthisapproach.Caffeineinhibits
Chk1 and sensitized mesothelioma cell-lines for pemetrexed, fur-
ther indicating that Chk1 suppression can increase response of an
antifolate agent and thus be a putative co-drug target for mesothe-
lioma [81]. CHEK1 is also an attractive target, as its protein is not
expressed in normal tissues except testis and tonsils, according to
the human protein atlas [82] and that it is selectively expressed in
mesothelioma tumour cells (Fig. 2). UCN-01 is a Chk1 inhibitor that
is currently in clinical trials.
Angiogenesis is important for tumour progression and sur-
vival [83]. VEGF is highly expressed in mesothelioma [84] but
was not differentially expressed here. On the contrary, AGGF1, a
recently discovered potent angiogenic [85] was differentially over-
expressed. This gene is implicated in the vascular overgrowth
syndrome Klippel–Treneunay and may be important in dam-
age response related to radiation defence, as ionizing radiation
induced overexpression of AGGF1 in lymphoblastoid cells [86].
VG5Q, the protein encoded by this gene was overexpressed in
>75% of tumour cells, also the sarcomatoid component, as well
as the endothelium of pathologic vessels. Suppression of this

Page 9

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
9
Fig. 5. Differentially expressed genes of leukocyte trans-endothelial migration, P<0.05 (in KEGG PATHWAYS, Kanehisa et al. [8]). Green are down-regulated, red are overex-
pressed. Eight of the 11 differentially expressed genes were down-regulated, mainly those that facilitate leukocyte migration.
pro-angiogenic protein has, to our knowledge, not been explored
in cancer, and could be an important adjunct to other treat-
ments.
NQO1overexpressionintumourisaprerequisitefortheeffectof
thenovelanti-cancercompoundbeta-Lapachonethatinduceselec-
tive tumour apoptosis by an unknown mechanism [87], as well as
radiosensitization in vitro. Moreover in cell lines NQO1 was found
to inhibit DNA polymerase alpha, DNA replication and thymidy-
late synthase activity, NFkappa-beta activity as well as induction of
topoisomerase II alpha mediated DNA breaks [76]. Several of the
genes behind those mechanisms are overexpressed in mesothe-
lioma, as well as NQO1 protein expression, and thus NQO1 may
be an attractive target of the novel compound beta-Lapachone.
RAD21 is a critical gene in double-strand DNA repair and mitotic
growth, but its relation to cancer has been unclear. RAD21 gene
overexpression was recently shown to be involved in the invasion
and metastasis in oral squamous cell carcinoma [88]. Silencing of
RAD21 gene expression decreased cell growth and enhanced cyto-
toxicity of etoposide and bleomycin in human breast cancer cells
[89], showing that it may serve as a novel target for developing
cancer therapeutics that can potentially enhance the anti-tumour
activity of chemotherapeutic agents acting via induction of DNA
damage.
4.12. Gene expression and survival
There have been attempts to connect gene expression in
mesothelioma to survival, but largely, the gene lists have been
incongruent [90]. Some of the genes overexpressed in our mate-
rial are included in the “death-from-cancer” signature predicting
therapy failure between patients [17]. This signature of eight over-
expressed and three down-regulated stem-cell related genes were
tested on several large microarray datasets with multiple tumour
types,includingmesothelioma[17].Weidentifiedthreeoftheeight
overexpressed genes, namely CCNB1, MKI67/Ki67 and BUB1. A 199
gene set from the same study also highlighted BIRK5/Survivin,
CCNB2, CDC2, TOP2A and CDC25 [91] as multi-cancer negative
prognostic factors, coinciding with our findings as well. Generally
these genes have all been linked to tumour aggressiveness and/or
survival that also fit very well with the mesothelioma phenotype.
There are several overexpressed genes in our material involved
in tumour aggressiveness and short survival (e.g. HSP90B1, ANX4,
Metadherin, ORAOV1, MAGED4, MTA3, ECT2, TSG101) where most
that deserve further study have not been described in mesothe-
lioma.
Three overexpressed genes found in mesothelioma versus nor-
maltissue(TYMS,RAD21,CHEK1)werecomparedbetweenthetwo
cases with the highest and lowest survival. We registered marked
differences where the most aggressive and resistant tumour had
both higher mRNA and protein expression (Fig. 6). The only long-
term survivor, case no. 8, had an expression profile more similar to
the controls, indicating a less perturbed genome than those with
short survival (data not shown) (Fig. 1). Our material is too small
to draw conclusions as to survival related to gene expression, but
these findings clearly warrant more gene-profiling studies with
more patient samples.

Page 10

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
10
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
Fig.6. Principalcomponentanalysisofonlytumoursamplesandsurvivalinmonths
(m).Threeoverexpressedgenesfoundinmesotheliomaversusnormaltissue(TYMS,
RAD21, CHEK1) are compared between the highest (upper row) versus the lowest
(lowerrow)survivalpatientaccordingtomRNAexpression(numbersbelowthehis-
tological pictures) and protein expression. Low TYMS gene expression and almost
absent protein expression correlated positively to survival and protracted antifolate
(pemetrexed) response. Low RAD21 and nuclear staining correlated to high sur-
vival. CHEK1 expression and protein staining are almost absent in normal tissues.
BothcontrolimportantDNA-repairfunctions,renderingthemattractivetherapeutic
targets (see text).
5. Conclusion
Mesothelioma resistance to therapy is reflected by its gene
profile where many known genes related to chemo- and radio-
resistance were significantly overexpressed. Specific suppression
of some genes or gene products that were discovered, could lead to
improved anti-tumour effect of conventional chemo- and/or radio-
therapy, notably microtubule related genes, cell cycle checkpoint
and DNA-repair genes, with the Fanconi anemia/BRCA2 pathway
as the most central. A very recent hypothesis claims that genomic
stabilization through DNA repair overexpression is a general pre-
requisite for tumour invasion and metastasis [92]. Our findings are
along this line as mesothelioma shares those characteristics. The
genes CHEK1, RAD21, BRCA2 AGGF1, NQO1 and RAN are proposed
as novel putative drug co-targets against mesothelioma.
Mesothelioma has been an unsolved oncological problem of
more than 40 years with therapeutic trials, often toxic with little
clinicalgain.Oneofthelessonsofthisfact,supportedbythepresent
study, is that malignant mesothelioma obviously has a heteroge-
nous and complex biology with versatile resistance mechanisms.
One should therefore be cautious not to start large patient studies
before testing drugs of interest in various combinations on several
preclinical models. There are indications that low expression of
target genes, as thymidylate syntase, is related to increased drug
sensitivity [93], an important point to bear in mind both for drug
designing and testing.
Conflict of interest statement
None.
Acknowledgements
The authors want to thank Unn Granli and Borgny Ytterhus for
excellentimmunohistochemistry.Wearemostgratefulforthecom-
ments of Prof. Hans Krokan.
The project was partially funded by the Cancer Foundation of St.
Olavs Hospital.
Microarray experiments were performed at the Norwegian
Microarray Consortium (NMC) at NTNU, Trondheim, which is sup-
ported by FUGE, The Norwegian Research Council.
The funding source has not influenced the conception, design or
analysis of this study.
References
[1] Robinson
2005;366:397–408.
[2] Bridda A, Padoan I, Mencarelli R, Frego M. Peritoneal mesothelioma: a review.
Med Gen Med 2007;9:32.
[3] Baas P. Chemotherapyformalignant
2005;49(Suppl. 1):S61–4.
[4] Vogelzang NJ, Rusthoven JJ, Symanowski J, Denham C, Kaukel E, Ruffie P, et
al. Phase III study of pemetrexed in combination with cisplatin versus cis-
platin alone in patients with malignant pleural mesothelioma. J Clin Oncol
2003;21:2636–44.
[5] SmythGK.Linearmodelsandempiricalbayesmethodsforassessingdifferential
expression in microarray experiments. Stat Appl Genet Mol Biol 2004;3 [article
3].
[6] Reiner A, Yekutieli D, Benjamini Y. Identifying differentially expressed
genes using false discovery rate controlling procedures. Bioinformatics
2003;19:368–75.
[7] Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and
powerful approach to multiple testing. J R Stat Soc 1995;57:289–300.
[8] KanehisaM,ArakiM,GotoS,HattoriM,HirakawaM,ItohM,etal.KEGGforlink-
inggenomestolifeandtheenvironment.NucleicAcidsRes2008;36:D480–484.
[9] Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, et al. Gene
ontology: tool for the unification of biology. The Gene Ontology Consortium.
Nat Genet 2000;25:25–9.
[10] Gidskehaug L, Anderssen E, Flatberg A, Alsberg BK. A framework for signifi-
cance analysis of gene expression data using dimension reduction methods.
BMC Bioinform 2007;8:346.
[11] BoutinC,ReyF.Thoracoscopyinpleuralmalignantmesothelioma:aprospective
study of 188 consecutive patients. Part 1: Diagnosis. Cancer 1993;72:389–93.
[12] Hassan R, Ho M. Mesothelin targeted cancer immunotherapy. Eur J Cancer
2008;44:46–53.
[13] Linden CJ, Mercke C, Albrechtsson U, Johansson L, Ewers SB. Effect of hemitho-
rax irradiation alone or combined with doxorubicin and cyclophosphamide
in 47 pleural mesotheliomas: a nonrandomized phase II study. Eur Respir J
1996;9:2565–72.
[14] Sorensen JB, Frank H, Palshof T. Cisplatin and vinorelbine first-line
chemotherapy in non-resectable malignant pleural mesothelioma. Br J Cancer
2008;99:44–50.
[15] McGrogan BT, Gilmartin B, Carney DN, McCann A. Taxanes, microtubules and
chemoresistant breast cancer. Biochim Biophys Acta 2008;1785:96–132.
[16] Yamamoto Y, Matsuyama H, Chochi Y, Okuda M, Kawauchi S, Inoue R, et al.
OverexpressionofBUBR1isassociatedwithchromosomalinstabilityinbladder
cancer. Cancer Genet Cytogenet 2007;174:42–7.
[17] Glinsky GV, Berezovska O, Glinskii AB. Microarray analysis identifies a death-
from-cancer signature predicting therapy failure in patients with multiple
types of cancer. J Clin Invest 2005;115:1503–21.
[18] Altieri DC. The case for survivin as a regulator of microtubule dynamics and
cell-death decisions. Curr Opin Cell Biol 2006;18:609–15.
[19] Xia F, Lee CW, Altieri DC. Tumor cell dependence on Ran-GTP-directed mitosis.
Cancer Res 2008;68:1826–33.
[20] Vaisberg EA, Koonce MP, McIntosh JR. Cytoplasmic dynein plays a role in mam-
malian mitotic spindle formation. J Cell Biol 1993;123:849–58.
[21] Suzuki SO, McKenney RJ, Mawatari SY, Mizuguchi M, Mikami A, Iwaki T, et al.
Expression patterns of LIS1, dynein and their interaction partners dynactin,
NudE, NudEL and NudC in human gliomas suggest roles in invasion and prolif-
eration. Acta Neuropathol 2007;113:591–9.
[22] Kim JC, Badano JL, Sibold S, Esmail MA, Hill J, Hoskins BE, et al. The Bardet-
Biedl protein BBS4 targets cargo to the pericentriolar region and is required for
microtubule anchoring and cell cycle progression. Nat Genet 2004;36:462–70.
[23] Marcus AI, Peters U, Thomas SL, Garrett S, Zelnak A, Kapoor TM, et al. Mitotic
kinesin inhibitors induce mitotic arrest and cell death in Taxol-resistant and
-sensitive cancer cells. J Biol Chem 2005;280:11569–77.
[24] Gurzov EN, Izquierdo M. RNA interference against Hec1 inhibits tumor growth
in vivo. Gene Ther 2006;13:1–7.
[25] Basto R, Gergely F, Draviam VM, Ohkura H, Liley K, Raff JW. Hsp90 is required
to localise cyclin B and Msps/ch-TOG to the mitotic spindle in Drosophila and
humans. J Cell Sci 2007;120:1278–87.
[26] Pespeni MH, Hodnett M, Abayasiriwardana KS, Roux J, Howard M, Broaddus
VC, et al. Sensitization of mesothelioma cells to tumor necrosis factor-related
apoptosis-inducing ligand-induced apoptosis by heat stress via the inhibi-
tion of the 3-phosphoinositide-dependent kinase 1/Akt pathway. Cancer Res
2007;67:2865–71.
[27] Lim ST, Mikolon D, Stupack DG, Schlaepfer DD. FERM control of FAK function:
implications for cancer therapy. Cell Cycle 2008;7:2306–14.
[28] Parsons JT, Slack-Davis J, Tilghman R, Roberts WG. Focal adhesion kinase: tar-
geting adhesion signaling pathways for therapeutic intervention. Clin Cancer
Res 2008;14:627–32.
BW,Musk AW,Lake RA.Malignant mesothelioma.Lancet
mesothelioma. LungCancer

Page 11

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
11
[29] Halder J, Landen Jr CN, Lutgendorf SK, Li Y, Jennings NB, Fan D, et al. Focal
adhesion kinase silencing augments docetaxel-mediated apoptosis in ovarian
cancer cells. Clin Cancer Res 2005;11:8829–36.
[30] Zhao R, Goldman ID. Resistance to antifolates. Oncogene 2003;22:7431–57.
[31] Hanauske AR, Eismann U, Oberschmidt O, Pospisil H, Hoffmann S, Hanauske-
Abel H, et al. In vitro chemosensitivity of freshly explanted tumor cells to
pemetrexed is correlated with target gene expression. Invest New Drugs
2007;25:417–23.
[32] Zheng Z, Li X, Schell MJ, Chen T, Boulware D, Robinson L, et al. Thymidylate
synthase in situ protein expression and survival in stage I nonsmall-cell lung
cancer. Cancer 2008;112:2765–73.
[33] Akbari M, Otterlei M, Pena-Diaz J, Aas PA, Kavli B, Liabakk NB, et al. Repair of
U/GandU/AinDNAbyUNG2-associatedrepaircomplexestakesplacepredom-
inantly by short-patch repair both in proliferating and growth-arrested cells.
Nucleic Acids Res 2004;32:5486–98.
[34] Berger SH, Pittman DL, Wyatt MD. Uracil in DNA: consequences for carcinogen-
esis and chemotherapy. Biochem Pharmacol 2008;76:697–706.
[35] Souglakos J, Boukovinas I, Taron M, Mendez P, Mavroudis D, Tripaki M, et
al. Ribonucleotide reductase subunits M1 and M2 mRNA expression levels
and clinical outcome of lung adenocarcinoma patients treated with doc-
etaxel/gemcitabine. Br J Cancer 2008;98:1710–5.
[36] Rasheed ZA, Rubin EH. Mechanisms of resistance to topoisomerase I-targeting
drugs. Oncogene 2003;22:7296–304.
[37] Cliby WA, Lewis KA, Lilly KK, Kaufmann SH. S phase and G2 arrests induced
by topoisomerase I poisons are dependent on ATR kinase function. J Biol Chem
2002;277:1599–606.
[38] Cusack Jr JC, Liu R, Houston M, Abendroth K, Elliott PJ, Adams J, et al. Enhanced
chemosensitivity to CPT-11 with proteasome inhibitor PS-341: implications for
systemic nuclear factor-kappaB inhibition. Cancer Res 2001;61:3535–40.
[39] Spencer DM, Bilardi RA, Koch TH, Post GC, Nafie JW, Kimura K, et al. DNA repair
in response to anthracycline-DNA adducts: a role for both homologous recom-
bination and nucleotide excision repair. Mutat Res 2008;638:110–21.
[40] Nakopoulou L, Lazaris AC, Kavantzas N, Alexandrou P, Athanassiadou P,
Keramopoulos A, et al. DNA topoisomerase II-alpha immunoreactivity as
a marker of tumor aggressiveness in invasive breast cancer. Pathobiology
2000;68:137–43.
[41] DiLeoA,ClaudinoWM,PestrinM,LicitraS,BiganzoliL.Usingspecificcytotoxics
with a targeted mind. Breast 2007;16(Suppl. 2):S120–6.
[42] Vang Nielsen K, Ejlertsen B, Moller S, Trost Jorgensen J, Knoop A, Knudsen H,
et al. The value of TOP2A gene copy number variation as a biomarker in breast
cancer: update of DBCG trial 89D. Acta Oncol 2008;47:725–34.
[43] Hillerdal GSJ, Sundström S, Vikström A, Hjerpe A. Treatment of malignant
pleural mesothelioma with liposomized doxorubicine: prolonged time to pro-
gression and good survival. A Nordic study. Clin Respir J 2008;2:5.
[44] Chawla SP, Legha SS, Benjamin RS. Detorubicin—an active anthracycline in
untreated metastatic melanoma. J Clin Oncol 1985;3:1529–34.
[45] Hillerdal G, Sundstrom S, Sorensen JB, for the Nordic Mesothelioma Group.
Malignant pleural mesothelioma treated with a combination of pegylated lipo-
somal doxorubicin, carboplatin and gemcitabine: The CCG study. Proc Am Soc
Clin Oncol 2003:22.
[46] Stewart DJ. Mechanisms of resistance to cisplatin and carboplatin. Crit Rev
Oncol Hematol 2007;63:12–31.
[47] Beall HD, Murphy AM, Siegel D, Hargreaves RH, Butler J, Ross D. Nicotinamide
adenine dinucleotide (phosphate): quinone oxidoreductase (DT-diaphorase)
as a target for bioreductive antitumor quinones: quinone cytotoxicity and
selectivity in human lung and breast cancer cell lines. Mol Pharmacol
1995;48:499–504.
[48] Hassan KA, Ang KK, El-Naggar AK, Story MD, Lee JI, Liu D, et al. Cyclin B1 over-
expression and resistance to radiotherapy in head and neck squamous cell
carcinoma. Cancer Res 2002;62:6414–7.
[49] AhmedKM,LiJJ.NF-kappaB-mediatedadaptiveresistancetoionizingradiation.
Free Radic Biol Med 2008;44:1–13.
[50] CapalboG,RodelC,StauberRH,KnauerSK,BacheM,KapplerM,etal.Theroleof
survivin for radiation therapy. Prognostic and predictive factor and therapeutic
target. Strahlenther Onkol 2007;183:593–9.
[51] Rabik CA, Dolan ME. Molecular mechanisms of resistance and toxicity associ-
ated with platinating agents. Cancer Treat Rev 2007;33:9–23.
[52] Mirchandani KD, D’Andrea AD. The Fanconi anemia/BRCA pathway: a coordi-
nator of cross-link repair. Exp Cell Res 2006;312:2647–53.
[53] Kennedy RD, D’Andrea AD. DNA repair pathways in clinical practice:
lessons from pediatric cancer susceptibility syndromes. J Clin Oncol
2006;24:3799–808.
[54] Chirnomas D, Taniguchi T, de la Vega M, Vaidya AP, Vasserman M, Hartman AR,
et al. Chemosensitization to cisplatin by inhibitors of the Fanconi anemia/BRCA
pathway. Mol Cancer Ther 2006;5:952–61.
[55] DurantS,KarranP.Vanillins—anovelfamilyofDNA-PKinhibitors.NucleicAcids
Res 2003;31:5501–12.
[56] Yu D, Sekine E, Fujimori A, Ochiya T, Okayasu R. Down regulation of BRCA2
causes radio-sensitization of human tumor cells in vitro and in vivo. Cancer Sci
2008;99:810–5.
[57] Marcou Y, D’Andrea A, Jeggo PA, Plowman PN. Normal cellular radiosensitiv-
ity in an adult Fanconi anaemia patient with marked clinical radiosensitivity.
Radiother Oncol 2001;60:75–9.
[58] Hartsuiker E, Vaessen E, Carr AM, Kohli J. Fission yeast Rad50 stimulates
sister chromatid recombination and links cohesion with repair. EMBO J
2001;20:6660–71.
[59] Jacquemont C, Taniguchi T. Proteasome function is required for DNA
damage response and fanconi anemia pathway activation. Cancer Res
2007;67:7395–405.
[60] Muto M, Fujimori A, Nenoi M, Daino K, Matsuda Y, Kuroiwa A, et al. Isolation
and characterization of a novel human radiosusceptibility gene, NP95. Radiat
Res 2006;166:723–33.
[61] Kauffmann A, Rosselli F, Lazar V, Winnepenninckx V, Mansuet-Lupo A, Dessen
P, et al. High expression of DNA repair pathways is associated with metastasis
in melanoma patients. Oncogene 2008;27:565–73.
[62] WangY,KlijnJG,ZhangY,SieuwertsAM,LookMP,YangF,etal.Gene-expression
profiles to predict distant metastasis of lymph-node-negative primary breast
cancer. Lancet 2005;365:671–9.
[63] Dyrskjot L, Thykjaer T, Kruhoffer M, Jensen JL, Marcussen N, Hamilton-Dutoit S,
et al. Identifying distinct classes of bladder carcinoma using microarrays. Nat
Genet 2003;33:90–6.
[64] Mekeel KL, Tang W, Kachnic LA, Luo CM, DeFrank JS, Powell SN. Inactiva-
tion of p53 results in high rates of homologous recombination. Oncogene
1997;14:1847–57.
[65] Hasan MK, Yaguchi T, Sugihara T, Kumar PK, Taira K, Reddel RR, et al. CARF
is a novel protein that cooperates with mouse p19ARF (human p14ARF) in
activating p53. J Biol Chem 2002;277:37765–70.
[66] Chan N, Koritzinsky M, Zhao H, Bindra R, Glazer PM, Powell S, et al. Chronic
hypoxia decreases synthesis of homologous recombination proteins to offset
chemoresistance and radioresistance. Cancer Res 2008;68:605–14.
[67] Ogiso Y, Tomida A, Lei S, Omura S, Tsuruo T. Proteasome inhibition circum-
vents solid tumor resistance to topoisomerase II-directed drugs. Cancer Res
2000;60:2429–34.
[68] Borczuk AC, Cappellini GC, Kim HK, Hesdorffer M, Taub RN, Powell CA.
Molecular profiling of malignant peritoneal mesothelioma identifies the
ubiquitin-proteasome pathway as a therapeutic target in poor prognosis
tumors. Oncogene 2007;26:610–7.
[69] Sartore-Bianchi A, Gasparri F, Galvani A, Nici L, Darnowski JW, Barbone D, et al.
Bortezomib inhibits nuclear factor-kappaB dependent survival and has potent
in vivo activity in mesothelioma. Clin Cancer Res 2007;13:5942–51.
[70] Baas P, Boogerd W, Dalesio O, Haringhuizen A, Custers F, van Zandwijk N.
Thalidomide in patients with malignant pleural mesothelioma. Lung Cancer
2005;48:291–6.
[71] Aragon-Ching JB, Li H, Gardner ER, Figg WD. Thalidomide analogues as anti-
cancer drugs. Recent Patents Anticancer Drug Discov 2007;2:167–74.
[72] Jackson DE. The unfolding tale of PECAM-1. FEBS Lett 2003;540:7–14.
[73] Wegmann F, Petri B, Khandoga AG, Moser C, Khandoga A, Volkery S, et al.
ESAM supports neutrophil extravasation, activation of Rho, and VEGF-induced
vascular permeability. J Exp Med 2006;203:1671–7.
[74] Ludwig RJ, Zollner TM, Santoso S, Hardt K, Gille J, Baatz H, et al. Junctional
adhesionmolecules(JAM)-Band-Ccontributetoleukocyteextravasationtothe
skinandmediatecutaneousinflammation.JInvestDermatol2005;125:969–76.
[75] KruklitisRJ,SinghalS,DelongP,KapoorV,StermanDH,KaiserLR,etal.Immuno-
gene therapy with interferon-beta before surgical debulking delays recurrence
and improves survival in a murine model of malignant mesothelioma. J Thorac
Cardiovasc Surg 2004;127:123–30.
[76] Pink JJ, Planchon SM, Tagliarino C, Varnes ME, Siegel D, Boothman DA.
NAD(P)H:Quinoneoxidoreductaseactivityistheprincipaldeterminantofbeta-
lapachone cytotoxicity. J Biol Chem 2000;275:5416–24.
[77] Nebert DW, Vasiliou V. Analysis of the glutathione S-transferase (GST) gene
family. Hum Genom 2004;1:460–4.
[78] Oscarson M, Zanger UM, Rifki OF, Klein K, Eichelbaum M, Meyer UA. Tran-
scriptional profiling of genes induced in the livers of patients treated with
carbamazepine. Clin Pharmacol Ther 2006;80:440–56.
[79] Bucher N, Britten CD. G2 checkpoint abrogation and checkpoint kinase-1 tar-
geting in the treatment of cancer. Br J Cancer 2008;98:523–8.
[80] GaoQ,ZhouJ,HuangX,ChenG,YeF,LuY,etal.Selectivetargetingofcheckpoint
kinase 1 in tumor cells with a novel potent oncolytic adenovirus. Mol Ther
2006;13:928–37.
[81] Min SH, Goldman ID, Zhao R. Caffeine markedly sensitizes human mesothe-
lioma cell lines to pemetrexed. Cancer Chemother Pharmacol 2008;61:819–27.
[82] Berglund L, Bjorling E, Oksvold P, Fagerberg L, Asplund A, Al-Khalili Szigyarto
C, et al. A gene-centric human protein atlas for expression profiles based on
antibodies. Mol Cell Proteomics 2008:20.
[83] OzdemirF,AkdoganR,AydinF,ReisA,KavgaciH,GulS,etal.TheeffectsofVEGF
and VEGFR-2 on survival in patients with gastric cancer. J Exp Clin Cancer Res
2006;25:83–8.
[84] OhtaY,ShridharV,BrightRK,KalemkerianGP,DuW,CarboneM,etal.VEGFand
VEGF type C play an important role in angiogenesis and lymphangiogenesis in
human malignant mesothelioma tumours. Br J Cancer 1999;81:54–61.
[85] Tian XL, Kadaba R, You SA, Liu M, Timur AA, Yang L, et al. Identification
of an angiogenic factor that when mutated causes susceptibility to Klippel-
Trenaunay syndrome. Nature 2004;427:640–5.
[86] LongXH,ZhaoZQ,HeXP,WangHP,XuQZ,AnJ,etal.Dose-dependentexpression
changesofearlyresponsegenestoionizingradiationinhumanlymphoblastoid
cells. Int J Mol Med 2007;19:607–15.
[87] Li Y, Sun X, LaMont JT, Pardee AB, Li CJ. Selective killing of cancer cells by beta-
lapachone: direct checkpoint activation as a strategy against cancer. Proc Natl
Acad Sci USA 2003;100:2674–8.
[88] Yamamoto G, Irie T, Aida T, Nagoshi Y, Tsuchiya R, Tachikawa T. Correlation of
invasion and metastasis of cancer cells, and expression of the RAD21 gene in
oral squamous cell carcinoma. Virchows Arch 2006;448:435–41.

Page 12

Please cite this article in press as: Røe OD, et al. Malignant pleural mesothelioma: Genome-wide expression patterns reflecting general
resistance mechanisms and a proposal of novel targets. Lung Cancer (2009), doi:10.1016/j.lungcan.2009.03.016
ARTICLE IN PRESS
G Model
LUNG-3311;No.of Pages12
12
O.D. Røe et al. / Lung Cancer xxx (2009) xxx–xxx
[89] AtienzaJM,RothRB,RosetteC,SmylieKJ,KammererS,RehbockJ,etal.Suppres-
sionofRAD21geneexpressiondecreasescellgrowthandenhancescytotoxicity
of etoposide and bleomycin in human breast cancer cells. Mol Cancer Ther
2005;4:361–8.
[90] Lopez-Rios F, Chuai S, Flores R, Shimizu S, Ohno T, Wakahara K, et al.
Global gene expression profiling of pleural mesotheliomas: overexpression
of aurora kinases and P16/CDKN2A deletion as prognostic factors and criti-
cal evaluation of microarray-based prognostic prediction. Cancer Res 2006;66:
2970–9.
[91] Glinsky GV. Genomic models of metastatic cancer: functional analysis
of death-from-cancer signature genes reveals aneuploid, anoikis-resistant,
metastasis-enabling phenotype with altered cell cycle control and acti-
vated Polycomb Group (PcG) protein chromatin silencing pathway. Cell Cycle
2006;5:1208–16.
[92] Sarasin A, Kauffmann A. Overexpression of DNA repair genes is associated with
metastasis: a new hypothesis. Mutat Res 2008;659:49–55.
[93] SeveP,DumontetC.IsclassIIIbeta-tubulinapredictivefactorinpatientsreceiv-
ing tubulin-binding agents? Lancet Oncol 2008;9:168–75.