ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Senior Assistant Jorma Arhippainen

Lecturer Santeri Palviainen

Professor Hannu Heusala

Professor Olli Vuolteenaho

Senior Researcher Eila Estola

Director Sinikka Eskelinen

Professor Jari Juga

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-9772-4 (Paperback)ISBN 978-951-42-9773-1 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1149

ACTA

Hanna-Leena R

onkainen

OULU 2012

D 1149

Hanna-Leena Ronkainen

NOVEL PROGNOSTIC BIOMARKERS FOR RENAL CELL CARCINOMA

UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU, FACULTY OF MEDICINE,INSTITUTE OF CLINICAL MEDICINE, DEPARTMENT OF SURGERY;INSTITUTE OF DIAGNOSTICS, DEPARTMENT OF PATHOLOGY

A C T A U N I V E R S I T A T I S O U L U E N S I SD M e d i c a 1 1 4 9

HANNA-LEENA RONKAINEN

NOVEL PROGNOSTIC BIOMARKERS FOR RENAL CELL CARCINOMA

Academic Dissertation to be presented with the assentof the Doctoral Training Committee of Health andBiosciences of the University of Oulu for public defencein Auditorium 1 of Oulu University Hospital, on 23March 2012, at 12 noon

UNIVERSITY OF OULU, OULU 2012

Copyright © 2012Acta Univ. Oul. D 1149, 2012

Supervised byDocent Markku VaaralaDocent Pasi HirvikoskiProfessor Ylermi Soini

Reviewed byProfessor Kimmo TaariDocent Paula Kujala

ISBN 978-951-42-9772-4 (Paperback)ISBN 978-951-42-9773-1 (PDF)

ISSN 0355-3221 (Printed)ISSN 1796-2234 (Online)

Cover DesignRaimo Ahonen

JUVENES PRINTTAMPERE 2012

Ronkainen, Hanna-Leena, Novel prognostic biomarkers for renal cell carcinoma. University of Oulu Graduate School; University of Oulu, Faculty of Medicine, Institute ofClinical Medicine, Department of Surgery; Institute of Diagnostics, Department of Pathology,P.O. Box 5000, FI-90014 University of Oulu, FinlandActa Univ. Oul. D 1149, 2012Oulu, Finland

Abstract

Background and aims: Stage and grade are the most widely used prognostic parameters for renalcell carcinoma (RCC). The clinical course of this disease is not, however, always predictable bytraditional prognostic factors. In the era of new molecular targeted therapies a more accurateprognostication of RCC patient survival is important for the individualization of treatment andfollow-up of patients. Despite exhaustive research there are still no prognostic biomarkers forRCC in clinical practice. In order to find novel prognostic tissue markers for RCC, we examinedthe expression of 14 biomarkers involved in carcinogenesis and clarified their prognosticsignificance in RCC.

Material and methods: Out of 189 consecutive patients who underwent surgery for kidneycancer at Oulu University Hospital in the 1990s, 152 patients with histologically verified RCCwere included in this study. The stage distribution was 70 (46%), 12 (8%), 51 (34%) and 19 (12%)patients with stages I-IV, respectively. The majority of the tumours (83 tumours, 55%) werenuclear grade II and 5 (3%), 40 (27%) and 22 (15%) of the tumours were grades I, III and IV,respectively. Clinical and follow-up data were obtained from patient records, the Finnish CancerRegistry and on demand from the Population Register Centre of Finland. The biomarkers studiedincluded markers of the oxidative and neuroendocrine systems as well as proteins related to celladhesion and migration, invasion, metastasis, inflammation and immune responses. Theexpression of various biomarkers was characterized via immunohistochemical tests of archivaltumour material. The staining intensity was compared to clinicopathological parameters andpatient RCC-specific survival.

Results: The 5-year RCC-specific survival was 77%. The expression of Toll-like receptor 9(TLR9) was an independent marker of favourable RCC-specific survival whereas cytoplasmicmyosin VI expression was found to be an independent prognostic factor of poor RCC-specificsurvival. Cell culture experiments showed how cyclooxygenase-2 (COX-2) expression isregulated by HuR in RCC. HuR and COX-2 immunoexpression were also related to decreasedRCC-specific survival. Immunostaining of Keap1 was associated with advanced RCC and amarker of a poorer RCC-specific prognosis. The expression of different neuroendocrine markerswas evaluated but we could not establish any prognostic value for them.

Conclusions: In particular, TLR9, HuR and myosin VI can be regarded as promising novelprognostic biomarkers in RCC. Stage, however, is the most important single prognostic factor forRCC.

Keywords: biological tumour markers, cyclooxygenase-2 (COX-2), HuR, Keap1,myosin VI, prognosis, renal cell carcinoma, survival, Toll-like receptor 9 (TLR9)

Ronkainen, Hanna-Leena, Uusia munuaissyövän ennusteellisia merkkiaineita. Oulun yliopiston tutkijakoulu; Oulun yliopisto, Lääketieteellinen tiedekunta, Kliinisenlääketieteen laitos, Kirurgia; Diagnostiikan laitos, Patologia, PL 5000, 90014 Oulun yliopistoActa Univ. Oul. D 1149, 2012Oulu

Tiivistelmä

Munuaissyöpä on vuosikymmenten ajan jatkuvasti yleistynyt. Vaikka se diagnosoidaan nykyisinuseimmiten sattumalöydöksenä vatsan alueen kuvantamistutkimuksissa ja hoitomenetelmät ovatviime vuosikymmenten aikana kehittyneet, munuaissyöpäkuolleisuus ei ole laskenut. Munuais-syövän ennusteen määrittäminen voi olla haasteellista. Perinteiset ennustetekijät, levinneisyys jaerilaistumisaste, eivät riitä selittämään kaikkien potilaiden taudinkulkua, eikä munuaissyövällevielä ole kliinisessä käytössä ennusteellista merkkiainetta. Munuaissyöpähoitojen kehittyessätaudinkulun ennustaminen on yhä tärkeämpää, jotta potilaiden hoito ja seuranta voidaan yksilöi-dä. Tämän väitöskirjatyön tarkoituksena oli etsiä uusia ennusteellisia kudosmerkkiaineita munu-aissyöpäkasvaimille.

Väitöskirjatutkimus perustuu 1990-luvulla Oulun yliopistollisessa sairaalassa leikatun 152munuaissyöpäpotilaan aineistoon. Lähes puolet aineiston kasvaimista edusti levinneisyysluok-kaa I, ja yli puolet munuaissyöpäkasvaimista oli hyvin erilaistuneita (tumagradus I ja II). Tutki-muspotilaista kerättiin kattavat seurantatiedot. Leikkauksessa poistettujen munuaissyöpäkasva-inten arkistomateriaalista tutkittiin eri merkkiaineiden ilmenemistä. Tutkitut merkkiaineet käsit-tivät oksidatiivisen ja neuroendokriinisen järjestelmän merkkiaineita sekä valkuaisaineita, jotkaliittyvät keskeisiin syövän ominaisuuksiin, kuten solujen välisiin liitoksiin ja solujen liikkumi-seen sekä etäpesäkkeiden syntymiseen. Lisäksi tutkittiin merkkiaineita, jotka liittyvät tulehdus-reaktioihin ja immuunipuolustukseen.

Väitöskirjatutkimus paljasti useita uusia kudosmerkkiaineita, joiden ilmeneminen munuais-syöpäkasvaimessa on yhteydessä potilaan ennusteeseen. Näistä merkittävimpiä ovat myosiiniVI, joka liittyy syöpäkasvainten metastasointiin, sekä immuunipuolustuksessa vaikuttava Tollinkaltainen reseptori 9 (Toll-like receptor 9, TLR9). Molemmat merkkiaineet osoittautuivat itse-näisiksi ennustetekijöiksi munuaissyövässä. Muita ennusteeseen vaikuttavia merkkiaineita ovattutkimuksen mukaan oksidatiivista stressiä aistiva Keap1 sekä immunologisiin reaktioihin liitty-vä syklo-oksigenaasi 2 (COX-2) ja sen ilmenemistä säätelevä HuR.

Asiasanat: biologiset kasvainmerkkiaineet, ennuste, HuR, Keap1, munuaissolu-karsinooma, syklo-oksigenaasi 2 (COX-2), TLR9, tyypin VI myosiini

7

Acknowledgements

This study was carried out at the Departments of Surgery and Pathology,

University of Oulu and Oulu University Hospital, during the years 2006–2011.

I want to express my deepest gratitude to my supervisors Docent Markku

Vaarala, Docent Pasi Hirvikoski and Professor Ylermi Soini. Markku introduced

me to this subject and gave me an opportunity to do scientific work. His

straightforward attitude has stepped up this study during times when research

work was not the first priority in my life. Pasi’s fatherly guidance and optimism

have been essential for completing this work. Ylermi’s expertise has carried this

research and his discerning comments and advice have been crucial for my

understanding of this work.

Professor Kimmo Taari and Docent Paula Kujala are acknowledged for their

critical appraisals of the manuscript of this thesis. I also wish to thank Professor

Jyrki Mäkelä and Docent Jukka Melkko who were monitors of my study.

I am deeply grateful to my superior, Docent Pekka Hellström, Head of

Division of Urology, who introduced me to the fascinating world of urology and

encouraged me to begin this scientific work.

I express my sincere gratitude to all of my co-authors. Docent Saila Kauppila

inspired me in the field of science and her practical approach helped in the

completion of this work. Professor Ari Ristimäki and Docent Katri S. Selander

kindly shared their experience and knowledge. Docent Katri Vuopala and Jenni

Rask, M.D., are also warmly thanked.

I want to thank the Head of Department of Operative Care, Docent Kari

Haukipuro, Professor Tatu Juvonen at the Department of Surgery and Professor

Tuomo Karttunen and Professor Markus Mäkinen at the Department of Pathology

who offered the research facilities for this study. Professor Timo Paavonen at the

Department of Pathology offered the research facilities and funding for this study

and he is also warmly thanked for his co-authorship. My warm thanks go to Mrs

Riitta Vuento, Mr Manu Tuovinen, Mrs Mirja Vahera, Mrs Erja Tomperi and Mrs

Mirja Mäkeläinen for their skilful technical assistance and friendly service

whenever I needed help or had questions, and Pasi Ohtonen, M.Sc., for his

instructive guidance and assistance with statistical analyses.

I want to thank my previous and present colleagues at the Department of

Surgery, who operated on and documented these patients during the 1990s. I wish

to express my warmest thanks to my colleagues and the personnel of the Division

of Urology for their kind and helpful attitude towards my work and for their

8

friendship and understanding during these years. Mrs Tuula Kivimaa is

acknowledged for her friendship and the practical help and encouragement I

received during this study. I feel gratitude for the invaluable assistance and

kindness I got over the years from the lately lamented Mrs Anita Tikkala.

I owe my heartfelt thanks to my friends who have not forgotten me during

these busy years. In particular, I want to thank my dear friend and colleague

Sanna Meriläinen, M.D., for the confidence boost and all the assistance and

advice she gave me.

My warmest thanks go to my parents, Kaarina and Ali Liikanen, to my sister

Sanna Liikanen and her family and to my parents-in-law Inkeri and Niilo

Ronkainen, whose help and love I have always been able to rely on. My aunt

Pirkko Soudunsaari and my godfather Erkki Hänninen are warmly thanked for

their assistance with child care and refreshing conversations during the writing

process.

Finally, I dedicate this work to my family: to my husband Jussi, who has

always encouraged and supported me and to our son Niilo, for being the light and

enjoyment in our lives. Jussi’s assistance with technical aspects and linguistic

problems has been invaluable.

Financial support from the Finnish Urological Association, the University of

Oulu, the Finnish Medical Foundation, the Cancer Society of Finland, the Cancer

Foundation of Northern Finland, and the Oulu University Hospital throughout

these years is greatly acknowledged.

Oulu February 2012 Hanna-Leena Ronkainen

9

Abbreviations

3’-UTR 3’ untranslated region (of mRNA)

5HT 5-hydroxytryptamine, serotonin

AEC 3-amino-9-ethylcarbazol

AKT protein kinase B

APC adenomatous polyposis coli, a tumour suppressor gene

ARE antioxidant response element or AU-rich element

B7 a family of immune-regulatory ligands

B7-H1 B7 homologue 1, programmed cell death 1 ligand 1, PD-L1, CD274,

an immunosuppressing protein

BHD Birt-Hogg-Dubé syndrome, folliculin (FLCN) tumour suppressor

gene

CAIX carbonic anhydrase IX

CD56 neural cell adhesion molecule, NCAM

CI confidence interval

c-MET mesenchymal-epithelial transition factor, a proto-oncogen

COX cyclooxygenase, prostaglandin endopedoxide H2 synthase, PGHS

COX-1 cyclooxygenase-1

COX-2 cyclooxygenase-2

CRP C-reactive protein

CT computed tomography

DAB 3–3'-diaminobenzidine

DNA deoxyribonucleic acid

E-cadherin epithelial cadherin, uvomorulin, L-CAM, cell-CAM 120/80, CD324

ELAV embryonic lethal abnormal vision, a family of RNA-binding

proteins

EMT epithelial to mesenchymal transition

FCS foetal calf serum

FH fumarate hydratase, a tumour suppressor gene

FLCN folliculin tumour suppressor gene

H & E haematoxylin and eosin

HGF hepatocyte growth factor

HIF hypoxia-inducible transcription factor

HR hazard ratio

HuR ELAVL1, HuA

IFN interferon

10

IGF insulin-like growth factor

IGFR1 type I insulin-like growth factor receptor

IL interleukin

IL-2R interleukin-2 receptor

Keap1 Kelch-like ECH-associated protein 1, BTB-Kelch-type substrate-

adaptor protein, INrf2

Ki-67 proliferation marker detected by the monoclonal antibody MIB-1,

MKI67

LEF lymphoid enhancer-binding factor, a transcription factor

MET mesenchymal to epithelial transition, mesenchymal-epithelial factor,

tyrosine kinase membrane receptor

miRNA microRNA

MMP matrix metalloproteinase

MRI magnetic resonance imaging

mRNA messenger RNA

mTOR mammalian target of rapamycin

MVI microscopic venous invasion

NCAM neural cell adhesion molecule, CD56

NF-kappaB nuclear factor kappa-light-chain-enhancer of activated B-cells, a

transcription factor

Nfr2 erythroid transcription factor NF-E2, nuclear factor erythroid-2-

related factor 2, NF-E2-related factor 2

NK natural killer cell

NSAIDs non-stereoidal anti-inflammatory drugs

NSE neurone-specific enolase, gamma-enolase, γ-enolase

NSS nephron-sparing surgery

p53 (tumour) protein 53, a tumour suppressor protein encoded by the

TP53 gene

PBS phosphate-buffered saline

PDGF platelet-derived growth factor

PG prostaglandin

Pl3K phosphatidylinositol-3-kinase

PSA prostate-specific antigen

PTEN phosphatase and tensin homologue, a tumour suppressor gene

PTHrP parathyroid hormone-related protein

pVHL von Hippel Lindau protein

RCC renal cell carcinoma

11

RNA ribonucleic acid

ROS reactive oxygen species

SD standard deviation

SDS sodium dodecyl sulphate, sodium lauryl sulphate

siRNA small interfering RNA

SPSS Statistical Package for the Social Sciences

T tumour class

TBS Tris-buffered saline

TCF T-cell factor, a transcription factor

TGF transforming growth factor

TLR Toll-like receptor

TLR9 Toll-like receptor 9

TMA tissue microarray

TNF tumour necrosis factor

TNM international tumour node metastasis classification by the UICC

TORC1 mTOR complex 1

TORC2 mTOR complex 2

Tris tris(hydroxymethyl)aminomethane

Tris-EDTA tris(hydroxymethyl)aminomethane-ethylenediaminetetraacetic acid

buffer

Trx thioredoxin, adult T-cell leukaemia-derived factor, ADF

UICC International Union Against Cancer

uPA urokinase-type plasminogen activator

US ultrasound examination

VEGF vascular endothelial growth factor

VEGFR vascular endothelial growth factor receptor

VHL von Hippel Lindau disease, von Hippel Lindau tumour suppressor

gene

Wnt wingless type

β-catenin beta-catenin

χ2 test chi-square test, chi-squared test

12

13

List of original publications

This thesis is based on the following articles, which are referred in the text by

Roman numerals. In addition, some unpublished data are presented.

I Ronkainen H, Vaarala MH, Kauppila S, Soini Y, Paavonen TK, Rask J & Hirvikoski P (2009) Increased BTB-Kelch type substrate adaptor protein immunoreactivity associates with advanced stage and poor differentiation in renal cell carcinoma. Oncol Rep 21(6): 1519–1523.

II Ronkainen H, Kauppila S, Hirvikoski P & Vaarala MH (2010) Evaluation of myosin VI, E-cadherin and beta-catenin immunostaining in renal cell carcinoma. J Exp Clin Cancer Res 29: 2.

III Ronkainen H*, Vaarala MH*, Hirvikoski P & Ristimäki A (2011) HuR expression is a marker of poor prognosis in renal cell carcinoma. Tumor Biol 32(3): 481–487.

IV Ronkainen H*, Hirvikoski P*, Kauppila S, Vuopala KS, Paavonen TK & Selander KS, Vaarala MH (2011) Absent Toll-like receptor 9 expression predicts poor prognosis in renal cell carcinoma. J Exp Clin Cancer Res 30: 84.

V Ronkainen H, Soini Y, Vaarala MH, Kauppila S & Hirvikoski P (2010) Evaluation of neuroendocrine markers in renal cell carcinoma. Diagn Pathol 5: 28.

*Contributed equally

14

15

Contents

Abstract

Tiivistelmä

Acknowledgements 7 Abbreviations 9 List of original publications 13 Contents 15 1 Introduction 19 2 Review of the literature 21

2.1 Epidemiology and aetiology ................................................................... 21 2.1.1 Incidence and prevalence ............................................................. 21 2.1.2 Risk factors ................................................................................... 21

2.2 Diagnosis ................................................................................................. 22 2.2.1 Clinical presentation ..................................................................... 22 2.2.2 Macroscopic tumour presentation ................................................ 23 2.2.3 Histopathology ............................................................................. 23 2.2.4 Laboratory examinations .............................................................. 24 2.2.5 Radiological evaluation ................................................................ 24

2.3 Tumour biology in renal cell carcinoma ................................................. 25 2.3.1 The hypoxia-inducible pathway ................................................... 25 2.3.2 The Pl3K/AKT/mTOR pathway ................................................... 26 2.3.3 The Wnt/beta-catenin pathway ..................................................... 27 2.3.4 Epithelial to mesenchymal transition ........................................... 27 2.3.5 The HGF/MET pathway ............................................................... 27 2.3.6 Other genetic alterations ............................................................... 28

2.4 Survival and prognostic factors in renal cell carcinoma ......................... 29 2.4.1 Clinical prognostic factors ............................................................ 29 2.4.2 Anatomical prognostic factors: stage ........................................... 30 2.4.3 Histological prognostic factors ..................................................... 32 2.4.4 Molecular prognostic markers ...................................................... 32 2.4.5 Prognostic factors in metastasized in renal cell carcinoma .......... 35 2.4.6 Prognostic systems and nomograms ............................................. 36

2.5 Treatment ................................................................................................ 37 2.5.1 Treatment of localized renal cell carcinoma ................................. 37 2.5.2 Treatment of metastasized renal cell carcinoma ........................... 38

2.6 Reasoning for studied biomarkers ........................................................... 38

16

2.6.1 Oxidative stress in carcinogenesis and markers of the

oxidative system ........................................................................... 38 2.6.2 Cell adhesion, migration, invasion and metastasis and

related proteins ............................................................................. 42 2.6.3 Inflammation-related cancer and the immunologic nature

of renal cell carcinoma ................................................................. 47 2.6.4 Neuroendocrine activity in cancer ................................................ 53

3 Aims of the study 57 4 Materials and methods 59

4.1 Study cohort ............................................................................................ 59 4.2 Clinical data and follow-up ..................................................................... 59 4.3 Tumour samples ...................................................................................... 60 4.4 Multi-tissue blocks .................................................................................. 60 4.5 Immunohistochemistry ............................................................................ 60 4.6 Evaluation of immunohistochemial staining ........................................... 62

4.6.1 Nfr2 (I) ......................................................................................... 62 4.6.2 Keap1 (I) ....................................................................................... 62 4.6.3 Thioredoxin (I) ............................................................................. 63 4.6.4 Myosin VI (II) .............................................................................. 63 4.6.5 E-cadherin (II) .............................................................................. 63 4.6.6 Beta-catenin (II) ............................................................................ 64 4.6.7 HuR (III) ....................................................................................... 64 4.6.8 Cyclooxygenase-2 (III) ................................................................. 64 4.6.9 Toll-like receptor 9 (IV) ............................................................... 64 4.6.10 Neuroendocrine markers (V) ........................................................ 64

4.7 Cell culture and RNA interference (III) .................................................. 65 4.8 Immunoblotting (III) ............................................................................... 65 4.9 Statistical analyses................................................................................... 66 4.10 Ethics ....................................................................................................... 66

5 Results 67 5.1 Characteristics, treatment and follow-up of the patients ......................... 67 5.2 Survival analysis of the clinicopathological characteristics .................... 68 5.3 Staining results and their association with the clinicopathological

parameters measured and disease-specific survival ................................ 71 5.3.1 Markers of the oxidative system (I) .............................................. 74 5.3.2 Myosin VI, E-cadherin and beta-catenin (II) ................................ 76 5.3.3 HuR and cyclooxygenase-2 (III) .................................................. 78

17

5.3.4 Toll-like receptor 9 (IV) ............................................................... 80 5.3.5 Markers of the neuroendocrine system (V) .................................. 81 5.3.6 The prognostic accuracy of tested biomarkers ............................. 82

6 Discussion 83 6.1 Material and methodology ...................................................................... 83 6.2 Patient survival and clinicopathological features as prognostic

factors ...................................................................................................... 84 6.3 Markers of the oxidative system (I) ........................................................ 85

6.3.1 Keap1 and Nfr2 ............................................................................ 85 6.3.2 Thioredoxin .................................................................................. 87

6.4 Myosin VI, E-cadherin and beta-catenin (II) .......................................... 87 6.4.1 Myosin VI ..................................................................................... 87 6.4.2 E-cadherin .................................................................................... 89 6.4.3 Beta-catenin .................................................................................. 89 6.4.4 Associations between myosin VI, E-cadherin and beta-

catenin .......................................................................................... 91 6.5 HuR and cyclooxygenase-2 (III) ............................................................. 91

6.5.1 HuR .............................................................................................. 92 6.5.2 Cyclooxygenase-2 ........................................................................ 94

6.6 Toll-like receptor 9 (IV) .......................................................................... 95 6.7 Markers of neuroendocrine activity (V) .................................................. 97 6.8 Challenges in searching for prognostic biomarkers for renal cell

carcinoma .............................................................................................. 100 6.9 Future aspects ........................................................................................ 102

7 Summary and conclusions 105 References 107 Original publications 135

18

19

1 Introduction

Renal cell carcinoma (RCC) is the most lethal urological malignancy, accounting

for 100,000 deaths worldwide annually (Parkin et al. 2005, Lam et al. 2008).

Approximately 20–30% of patients are diagnosed at the metastatic stage of

disease and half of the remaining patients will experience recurrence after an

initially curative treatment (Motzer et al. 1996, Crispen et al. 2008, Lam et al. 2008). Predicting the clinical outcome of individual RCC patients is challenging

and not always possible with classic prognostic factors, staging and grading,

which are primarily used when assessing cancer prognosis (Crispen et al. 2008,

Volpe & Patard 2010). During the last decade the treatment of metastatic RCC has

dramatically changed, giving new hope to patients suffering from this malignancy,

where survival has traditionally been regarded as poor in advanced stages. In the

era of new molecular targeted therapies there is a definite need for better tools to

predict the clinical course of RCC. Accurate prognostication would help in patient

counselling and to plan and individualize patient treatment and follow-up. (Volpe

& Patard 2010) High-risk patients could be selected for more effective treatments,

more careful surveillance and clinical trials with adjuvant therapies. On the other

hand, patients with indolent disease could be spared from over-treatment,

psychological stress and adverse effects of follow-up, such as radiation exposure,

which would also save health care costs and resources. (Eichelberg et al. 2009,

Volpe & Patard 2010)

Biomarkers are objectively measured and evaluated indicators of biological

or pathological processes or treatment responses. Cancer biomarkers are typically

produced by the tumour or by the body in response to the tumour. Prognostic

biomarkers are used to categorize patients into different risk groups and to predict

the course of a disease. (Bensalah et al. 2007) The ideal prognostic biomarker

provides prognostic information that is not provided by available

clinicopathological indices. Other criteria for prognostic biomarkers are statistical

significance, reproducibility, standardization, external validation and feasibility,

such as suitability for daily clinical practices, for example evaluations based on

urine and blood samples, and reasonable costs. (Crispen et al. 2008)

Immunohistochemical techniques are used to determine the expression and

cellular and subcellular locations of proteins of interest in tissue samples. In the

tissue microarray technique (TMA), multiple samples are collected in the same

block, which allows the high-throughput analysis of hundreds of specimens at the

20

same time in a cost-effective manner. (Merseburger et al. 2006, Di Napoli &

Signoretti 2009)

Renal cell carcinoma is recognized as a family of cancers that originate from

the renal tubular epithelium and have distinct genetic and molecular backgrounds,

unique morphological features and a characteristic clinical course (Lam et al. 2008). The advances in gene technology, molecular biology and proteomics have

enabled the identification of underlying molecular pathways of RCC as well as

therapeutic targets for advanced disease (Pfaffenroth & Linehan 2008, Finley et al. 2011). Thousands of genes and a large number of molecular biomarkers have

been screened and tested in order to find a prognostic biomarker for RCC. Despite

exhaustive research, none of the markers tested so far have shown promise as an

independent prognostic factor that could improve the predictive accuracy of

current prognostic systems and be suitable for clinical practice. (Nogueira & Kim

2008, Volpe & Patard 2010)

Malignant transformation from normal to neoplastic cells involves acquired

gain-of-function mutations in oncogenes and loss-of-function mutations in tumour

suppressor genes and genome maintenance genes (Vogelstein & Kinzler 2004).

Together, these provide growth advantages to tumour cells that are connected to

the characteristics of cancer, such enhanced cell division (proliferation), the

evasion of growth suppressors, resistance to apoptosis, induction of angiogenesis,

invasion and metastasis, reprogramming of energy metabolism and the avoidance

of antitumoural immune responses. Genomic instability is a fundamental

underlying mechanism in cancer that provides genetic diversity, which further

enhances carcinogenesis. In addition, inflammation promotes multiple hallmarks

of cancer. (Hanahan & Weinberg 2011)

The present study was designed to evaluate the expression and prognostic

value of novel biomarkers in RCC in order to find prognostic biomarkers for

clinical use. The markers selected contribute to the basic hallmarks of cancer,

such as cell proliferation, the inhibition of apoptosis, invasion and metastasis and

angiogenesis. The markers included proteins involved in the oxidative system,

cell adhesion and cell migration, inflammation and immune surveillance and

neuroendocrine differentiation. The markers were studied using the

immunohistochemical TMA technique in archival RCC tumour materials

collected from patients who were operated in the 1990s at Oulu University

Hospital. The immunoexpression of the biomarkers was assessed in context with

various clinicopathological features of RCC tumours and compared to patient

RCC-specific survival.

21

2 Review of the literature

2.1 Epidemiology and aetiology

2.1.1 Incidence and prevalence

Renal cell carcinoma accounts for approximately 85% of malignant kidney

tumours and 2% of all malignant neoplasms (Motzer et al. 1996). The highest

incidence rates occur in Europe and North America and the lowest in Asia and

Africa. Among females, the incidence rates are generally about one half of those

observed among males. (Lipworth et al. 2006, Chow et al. 2010) At the time of

diagnosis the average age of patients is in the early sixties (Lipworth et al. 2006).

The incidence of RCC is 8.9 per 100,000 persons per year, with the peak

incidence occuring between 60 and 70 years of age (Campbell et al. 2007).

During the last few decades the incidence and mortality of RCC have shown a

continual increase. The increase in incidence can partly be explained by

incidentally discovered, smaller and more often localized tumours. In recent years,

however, the incidence and mortality rates of RCC have stabilized worldwide.

(Chow et al. 2010) In Finland, approximately 900 new kidney cancers are

diagnosed and over 300 RCC patients die of their disease each year. In 2010,

there were over 6500 kidney cancer patients in Finland. (Finnish Cancer Registry

2011)

2.1.2 Risk factors

The most common and generally recognized risk factors for RCC are obesity and

cigarette smoking, accounting for 30% and 20% of RCCs, respectively.

Hypertension increases the risk of RCC, while antihypertensive drugs such as

diuretics are not independent risk factors. Acquired renal cystic disease and end-

stage renal disease are associated with an increased risk of RCC. (Lipworth et al. 2006, Chow et al. 2010) Ionizing radiation weakly increases the risk of RCC

(Lipworth et al. 2006). Diets containing a large amount of fruit and vegetables, a

lower caloric intake as well as physical activity and moderate alcohol

consumption may have a protective effect against RCC (Chow et al. 2010).

Exposure to different occupational agents such trichloroethylene (TCE) and

asbestos is an ambiguous risk factor for RCC and, generally, RCC is not regarded

22

as an occupational disease (Lipworth et al. 2006, Chow et al. 2010). Furthermore,

there is no convincing evidence to suggest that non-stereoidal anti-inflammatory

drugs (NSAIDs), largely salicylates and acetaminophen, are risk factors for RCC

(Lipworth et al. 2006).

A self-reported family history of RCC is associated with a 2–3-fold increase

in RCC risk and it is believed that genetic factors associate with environmental

factors to increase the risk of RCC (Lipworth et al. 2006, Chow et al. 2010).

2.2 Diagnosis

2.2.1 Clinical presentation

A renal mass can grow rather large in the retroperitoneum, until it becomes

symptomatic and palpable. Symptoms related to RCC are due to the local growth

of the tumour, haemorrhage, paraneoplastic syndromes or metastasis. (Campbell et al. 2007) The most common presentations of RCC are hematuria, abdominal

pain and a palpable mass in the flank or abdomen. The classic triad of flank pain,

gross hematuria and a palpable abdominal mass is found in less than 10% of cases.

A varicocele, usually left-sided, resulting from the obstruction of the testicular

vein, is present in 2% of male patients. (Motzer et al. 1996) Paraneoplastic

syndromes result from the humoral release of various tumour-associated proteins,

which are directly produced by the cancer cells or by the immune system in

response to the tumour. Paraneoplastic symptoms and signs include, for example,

hypertension, elevated erythrocyte sedimentation rate, anaemia, polycythemia,

cachexia, weight loss, malaise, pyrexia, hypercalcaemia, neuromyopathy,

amyloidosis, hypoalbuminemia or hepatic dysfunction and occur in 20% of RCCs.

(Campbell et al. 2007, Lam et al. 2008) Patients can suffer from symptoms due to

metastases, such as persistent cough and bone pain. The most frequent sites for

RCC metastases are lung parenchyma, bone, liver and brain, but RCCs can

metastasize to virtually any organ site. (Motzer et al. 1996, Ljungberg et al. 2010)

The role of a physical examination is limited in the diagnosis of RCC but it might

provide clues about a more advanced stage of disease, such as venous

involvement and metastases (Ljungberg et al. 2010).

23

2.2.2 Macroscopic tumour presentation

Macroscopically, RCC usually appears as a solid round or ovoid lesion

circumscribed by a pseudocapsule. Cystic degeneration and calsification are

found in 10–25% and 10–20% of the tumours, respectively. One unique feature of

RCC is its capability to grow into the venous system. Most sporadic RCCs are

unilateral and unifocal, whereas bilateral and multicentral involvement is more

common with the familial form of RCC. (Campbell et al. 2007)

2.2.3 Histopathology

Renal cell carcinomas originate from the renal tubular epithelium and have

distinct genetic abnormalities and unique morphological features (Kovács et al. 1997, Lam et al. 2008). The histological diagnosis of RCC is usually based on

morphological assessment under a microscope. When histological findings

overlap, immunohistochemical and microRNA (miRNA) techniques can be

helpful in distinguishing different subtypes of RCC. (Kovács et al. 1997, Kim &

Kim 2002, Eble 2004, Youssef et al. 2011) The nuclear grading of RCC is mostly

based on the Fuhrman grading system, which is a four-tiered grading system that

takes into account nuclear and nucleolar size, shape and content (Fuhrman et al. 1982).

Classification of renal cell carcinomas

Clear cell carcinoma accounts for approximately 70–80% of RCCs. The most

frequently recognized genetic abnormalities in clear cell RCC are mutations of

the von Hippel Lindau (VHL) gene, the duplication of chromosome 5q and

deletions at chromosomes 3p, 6q, 8p, 9p and 14q. The histological picture of clear

cell RCC is characterized by the predominance of cells with a clear cytoplasm,

although there can also be foci with an eosinophilic cytoplasm. (Kovács et al. 1997, Campbell et al. 2007)

Papillary carcinoma represents 10–15% of RCCs and it has a tendency

towards multicentricity. Microscopically, basophilic and eosinophilic cells are

arranged in papillary or tubular configurations. Type 1 papillary RCC consists of

basophilic cells with scant cytoplasm, whereas the potentially more aggressive

type 2 consists of eosinophilic cells and an abundant granular cytoplasm. (Eble et al. 2004, Campbell et al. 2007) Papillary RCCs are marked by the loss of Y

24

chromosomes in males and trisomy of chromosomes 3q, 7, 8p, 12, 16, 17 and 20.

(Kovács et al. 1997).

Chromophobic carcinoma accounts for 3–5% of RCCs and it is mostly

sporadic, although some cases are related to the Birt-Hogg-Dubé (BHD)

syndrome (Eble et al. 2004, Campbell et al. 2007). Microscopically, the cells are

characterized by a pale or eosinophilic granular cytoplasm. The underlying

genetic abnormality is monosomy in chromosomes 1, 2, 6, 10, 13, 17 or 21.

(Kovács et al. 1997)

Collecting duct or Bellini’s duct carcinoma represents less than 1% of RCCs

(Eble et al. 2004, Campbell et al. 2007). It is characterized by irregular channels

lined by a highly atypical epithelium, which can have a hobnail appereance

(Kovács et al. 1997). Renal medullary carcinoma is considered a subtype of

collecting duct carcinoma and associated with the sickle cell trait. Less than 3%

of RCCs are characterized by indeterminant histological features and they are

called unclassified or undifferentiated RCCs. (Campbell et al. 2007)

2.2.4 Laboratory examinations

The most commonly used laboratory parameters in RCC are C-reactive protein

(CRP), haemoglobin, erythrocyte sedimentation rate, alkaline phosphatase, serum

calcium, serum creatinine and glomerular filtration rate. A more precise

evaluation of renal functioning, such as by an isotope renogram, is particularly

important for treatment decisions when the tumour involves a solitary kidney,

when the tumours are bilateral, when renal function is decreased or if there is a

risk of renal impairment in the future. (Ljungberg et al. 2010)

2.2.5 Radiological evaluation

Over one half of RCCs are detected incidentally in radiological examinations

(Pantuck et al. 2000). A renal mass observed by ultrasound examination (US) will

be investigated further by an enhanced computed tomography (CT) scan.

Preoperative staging includes chest and abdominal CT scans. When vena caval

involvement is suspected, magnetic resonance imaging (MRI) or Doppler

ultrasound examination may provide further information. (Heidenreich et al. 2004,

Ljungberg et al. 2010) Angiography and embolization may be indicated

preoperatively for large tumours in order to reduce intraoperative bleeding or as

palliative management for pain and hemorrhage (Heidenreich et al. 2004). Further

25

examinations such as a bone scan and brain MRI or CT are performed when

metastases are suspected. A needle biopsy under image-guidance is performed

when a histological diagnosis is needed before oncological or minimally invasive

treatment or surveillance is considered. (Ljungberg et al. 2010)

2.3 Tumour biology in renal cell carcinoma

The majority of RCCs are sporadic; inherited forms represent only about 2% of

RCCs (Lipworth et al. 2006). However, current knowledge about the underlying

genetics and molecular biology of sporadic RCCs is predominantly based on

work with familial RCC (Linehan et al. 2003, Pfaffenroth & Linehan 2008).

2.3.1 The hypoxia-inducible pathway

Angiogenesis is an important part of the growth, invasive progression and

metastatic spread of solid tumours (Díaz-Flores et al. 1994). Hypoxia and the

compensatory hyperactivation of angiogenesis are important in the pathogenesis

of RCC, which is typically a highly vascularized neoplasm (Kaelin 2002, Ruan et al. 2009). The most commonly inherited form of RCC is associated with von

Hippel-Lindau (VHL) disease, which is an autosomal dominantly inherited

disorder caused by germline mutations in the VHL tumour suppressor gene. In

addition to multiple, bilateral clear cell RCCs, VHL disease is characterized by

haemangioblastomas of the retina and central nervous system,

phaeochromosytomas and renal, pancreatic and epididymal cysts. (Maher et al. 1991, Gnarra et al. 1994) The VHL gene resides on chromosome 3p25 (Latif et al. 1993) and its inactivation by somatic mutations or promoter hypermethylation is

also a common feature in sporadic clear cell RCCs, being present in 57% and

19% cases, respectively (Gnarra et al. 1994, Herman et al. 1994).

The VHL gene encodes the VHL protein (pVHL), which shuttles between the

nucleus and cytoplasm and is found in various cellular compartments such as the

endoplasmic reticulum and mitochondia (reviewed in Kaelin 2002). The VHL

protein is a component of a multiprotein complex called the E3 ubiquitin-ligase

complex that targets hypoxia-inducible transcription factors (HIFs) for ubiquitin-

mediated degradation. In normal cells and under normoxic conditions, HIFs are

hydroxylated and then bound by pVHL and targeted for ubiquitin-mediated

proteolysis. Under hypoxia, unhydroxylated HIFs avoid detection by pVHL and

accumulate. (Maxwell et al. 1999, reviewed in Kaelin 2004) The accumulation of

26

HIF also takes place as a consequence of the lack of pVHL. Hypoxia-inducible

transcription factors are heterodimers consisting of an α-subunit and a β-subunit.

Their accumulation results in the transcription of various genes involved in

adaptation to hypoxia. The target genes of HIF include, for example, vascular

endothelial growth factor (VEGF), which is involved in angiogenesis, and

transforming growth factor alpha (TGF-α) and platelet-derived growth factor

(PDGF), which contribute to mitogenesis. In addition to angiogenesis and cell

cycle control, the loss of functioning VHL has been implicated in various other

central processes of carcinogenesis, including differentiation and extracellular

matrix formation and turnover. (reviewed in Kaelin 2002, Kaelin 2004,

Pfaffenroth & Linehan 2008)

2.3.2 The Pl3K/AKT/mTOR pathway

Protein kinase B (AKT) and the mammalian target of rapamycin (mTOR) have a

central role in signalling pathways that drive tumorigenesis (Shaw & Cantley

2006). The growth factors VEGF and PDGF bind to their receptor tyrosine

kinases and activate phosphatidylinositol-3-kinase (Pl3K) and AKT. Activated

AKT regulates cell survival, growth and metabolism, as well as cell-cycle

progression, and leads to the inhibition of apoptosis, the accumulation of cell

cycle promoters and the activation of mTOR. (Shaw & Cantley 2006, reviewed in

Banumathy & Cairns 2010) The suggested underlying mechanism of AKT

activation in RCC is the negative regulation by the tumour suppressor gene called

the phosphatase and tensin homologue (PTEN). However, the genetic deficiences

in the Pl3K/AKT/mTOR pathway that cause RCC are still unclear. (reviewed in

Banumathy & Cairns 2010)

The mTOR kinase is an intracellular serine/threonine kinase that consists of

two distinct complexes: TORC1 and TORC2. The TORC1 complex regulates

protein synthesis and controls various specific cell growth regulators, such as the

transcription factor HIF-1α, that link mTOR signalling to diverse oncogenic

processes like angiogenesis. In turn, the TORC2 complex controls cell polarity

and spatial growth. The activity of mTOR is regulated by various growth factors,

the availability of nutritients and AKT, which in turn is also activated by mTOR.

(Shaw & Cantley 2006, reviewed in Banumathy & Cairns 2010)

27

2.3.3 The Wnt/beta-catenin pathway

The wingless-type (Wnt) signalling transduction pathway is a network of separate

but interacting pathways that have a central role in the pathogenesis of RCC

(reviewed in Karim et al. 2004, Banumathy & Cairns 2010). Several components

of the Wnt pathway are associated with carcinogenesis, especially beta-catenin

and the adenomatous polyposis coli (APC) tumour suppressor gene (see section

2.6.2). The signalling of Wnt positively regulates beta-catenin by inhibiting its

phosphorylation, ubiquitination and degradation. The tumour suppressor gene

APC in turn facilitates beta-catenin proteosomal degradation. The cytoplasmic

accumulation of beta-catenin results in the translocation of beta-catenin to the

nucleus, where it regulates gene expression by direct interaction with

transcription factors of the TCF/LEF family (T-cell factor, lymphoid enhancer-

binding factor). (Behrens et al. 1996, reviewed in Karim et al. 2004) The eventual

outcome of Wnt signalling is the transcription and expression of target genes

involved in the proliferation, differentiation, apoptosis and oncogenic

transformation and progression. Furthermore, Wnt signalling may also activate

the mTOR pathway (see section 2.3.2). (reviewed in Karim et al. 2004,

Banumathy & Cairns 2010)

2.3.4 Epithelial to mesenchymal transition

In the embryonic development of the kidney, mesenchymal to epithelial transition

is needed for organogenesis. The reverse phenomenon, epithelial to mesenchymal

transition (EMT), is a complex process that involves various signalling receptors,

transcriptional regulators and target genes and is essential before clear cell RCC

can metastasize. The main target of these regulators is E-cadherin, which is

crucial for maintaining an epithelial phenotype and also regulates the expression

of beta-catenin, which is involved in Wnt signalling (see sections 2.3.3 and 2.6.2).

(reviewed in Banumathy & Cairns 2010)

2.3.5 The HGF/MET pathway

The mesenchymal-epithelial transition factor (c-MET) is a proto-oncogene that

encodes a tyrosine kinase membrane receptor (MET receptor), the ligand of

which is the hepatocyte growth factor (HGF). Both the c-MET gene and the gene

encoding HGF are located on chromosome 7. Ligand binding to the MET

28

receptor launches a signalling cascade with multiple events involving, for

example, proliferation, survival, motility and the morphogenesis of many cell

types. Following a gain-of-function mutation the MET receptor is constitutively

activated, leading to a dysregulated tumourigenic state characterized by

unregulated proliferation, transformation and increased invasive potential of the

cells. (reviewed in Pfaffenroth & Linehan 2008, Banumathy & Cairns 2010,

Finley et al. 2011) It has also been established that the loss of VHL promotes

oncogenic signalling via HGF (Peruzzi et al. 2006). The germline mutations of c-

MET are responsible for the hereditary papillary RCC (HPRCC) syndrome

characterized by type 1 papillary RCC (Schmidt et al. 1997). In addition,

activating mutations of c-MET are found in 5–13% of sporadic papillary RCCs,

although the primary underlying genetic abnormality in sporadic papillary RCC is

trisomy of chromosome 7, which is found in approximately 75% of cases

(reviewed in Pfaffenroth & Linehan 2008).

2.3.6 Other genetic alterations

Hereditary leiomyomatosis and renal cell carcinoma (HLRCC) is characterized by

germline mutations in the tumour suppressor gene fumarate hydratase (FH) and

papillary type 2 RCC (Linehan et al. 2003). Thus far, no relevant evidence has

been found to suggest somatic mutations of FH in sporadic RCC (reviewed in

Pfaffenroth & Linehan 2008). A germline mutation in the tumour suppressor gene

BHD (folliculin gene, FLCN gene) is associated with Birt-Hogg-Dubé (BHD)

syndrome that is characterized by renal tumours with variable histology, including,

for example, chromophobic and clear cell RCC. It has been suggested that

folliculin is a downstream effector of mTOR signalling. (reviewed in Pfaffenroth

& Linehan 2008) In addition, familial RCC syndromes have been shown to be

associated with the chromosome 3;8 translocation (Li et al. 1993, Gnarra et al. 1994), tuberous sclerosis complex genes TSC1 and TSC2 and the succinate

hydrogenase (SDH) gene (reviewed in Banumathy & Cairns 2010). Increased risk

of RCC has also been reported in patients with hyperparathyroidism-jaw tumour

syndrome (Haven et al. 2000) and hereditary nonpolyposis colorectal carcinoma

(Baiyee & Banner 2006).

29

2.4 Survival and prognostic factors in renal cell carcinoma

The clinical course of RCC is variable. Small, incidentally discovered tumours

can have indolent course even without treatment and in metastasized or recurrent

disease survival rates have traditionally been poor. The prognosis of RCC has

generally improved, which can be explained by early diagnosis and advances in

imaging, staging and treatment, including both surgery and medical therapy.

(Pantuck et al. 2001, Crispen et al. 2008) The most important prognostic factors

for RCC are the stage and grade (Volpe & Patard 2010).

2.4.1 Clinical prognostic factors

The presence of clinical symptoms is associated with decreased survival, whereas

incidentally found tumours seem to have a better prognosis, which is probably

related to smaller size and lower stage of the tumours (Pantuck et al. 2000,

Méjean et al. 2003, Schips et al. 2003). Weight loss of more than 10% of body

weight during 6 months is related to decreased survival and cachexia is an

independent marker of poor prognosis (Méjean et al. 2003, Lam et al. 2008). The

performance status, as assessed by the Eastern Cooperative Oncology Group

(ECOG) or Karnofsky scales, which indicate the impact of disease on the overall

health of the patient, have become established and significant prognostic markers

for RCC (Tsui et al. 2000, Lam et al. 2008). Younger age has been shown to be an

independent marker of a more favourable prognosis (Lang et al. 2004), whereas

gender does not have any prognostic potential (Schips et al. 2003).

Many laboratory parameters correlate with patient survival. In advanced RCC,

high CRP caused by excessive interleukin-6 (IL-6) production, which is a

multifunctional cytokine with growth factor activities, is associated with an

unfavourable prognosis. An elevated erythocyte sedimentation rate and

thrombocytosis are also correlated with poor patient outcome. (Lam et al. 2008)

In addition, serum calcium, haemoglobin, albumin, lactate dehydrogenase,

alkaline phophatase and neurone-specific enolase (NSE, γ-enolase) have at least

some prognostic value in RCC (Takashi et al. 1989, Rasmuson et al. 1993,

Méjean et al. 2003).

30

2.4.2 Anatomical prognostic factors: stage

The cancer stage is the most reliable prognostic factor in RCC. Generally, 5-year

RCC-specific survival rates after radical nephrectomy are 75–95% for localized

disease, 65–80% for locally advanced disease, 40–60% for tumours with vena

cava thrombus, 10–20% for lymph node involvement and 0–5% for metastatic

RCC. (Lam et al. 2008) Staging integrates the anatomical features of a tumour

including tumour size, renal capsule and venous invasion and adrenal

involvement and the presence of lymph node and distant metastasis which all are

well-known prognostic factors and, when assessed together, provide the most

reliable prognostic information in RCC (Delahunt et al. 2007, Ljungberg et al. 2010, Volpe & Patard 2010). The most commonly used staging systems for RCC

are the Tumour Node Metastasis (TNM) classification by the International Union

Against Cancer (UICC), presented in Figure 1 (Sobin & Wittekind 2002), and the

Robson classification, which is mostly used in the United States (Robson et al. 1969). The TNM scores can be further combined in stages I-IV (Figure 1). The

TNM classification was recently revised in 2009 (Sobin et al. 2009). Compared to

the previous version from the year 2002 (Sobin & Wittekind 2002), tumour class

T2 is now divided into subclasses T2a and T2b, including tumours that are more

than 7 cm but less than 10 cm in diameter and tumours more than 10 cm in

diameter, both limited to the kidney, respectively. In addition, T3a now includes

tumours with a tumour thrombus extending into the renal vein only and adrenal

gland invasion is now classified as T4. (Ljungberg et al. 2010)

31

Fig

ure

1.

TN

M c

las

sif

ica

tio

n o

f re

na

l c

ell

ca

rcin

om

a b

y t

he

UIC

C (

20

02

) F

rom

: E

ble

JN

et.

al.

20

04

. R

ep

rod

uc

ed

wit

h p

erm

iss

ion

by

IAR

C,

Ly

on

.

32

2.4.3 Histological prognostic factors

Despite criticism concerning the validity and predictive value of the Fuhrman

grading system, it is the most accurate histological grading system for RCC and

an independent prognostic factor for clear cell RCCs at least (Lang et al. 2004,

Delahunt et al. 2007, Volpe & Patard 2010). The 5-year RCC-specific survival is

89%, 65% and 46% for grades I, II and III-IV, respectively (Tsui et al. 2000).

Generally, the prognosis for papillary and chromophobic RCCs is better than

that of clear cell RCC, and the survival rate of collecting duct RCC is poor. After

the stratification of tumour stage, however, the prognostic information regarding

the histological subtype is lost. (Ljungberg et al. 1999, Delahunt et al. 2007,

Ljungberg et al. 2010, Volpe & Patard 2010) Among papillary RCCs, type 1

tumours have a better prognosis than type 2, which are independent markers of a

poor prognosis (Delahunt et al. 2001). Out of the various histological features of

RCC, microscopic venous invasion (MVI), sarcomatoid differentiation, necrosis

and invasion of the collection system are associated with reduced survival rates,

whereas cystic composition is related to a more benign clinical course (Lang et al. 2004, Delahunt et al. 2007, Ljungberg et al. 2010).

2.4.4 Molecular prognostic markers

Although numerous biomarkers have been recognized as being associated with

patient outcome, no prognostic biomarker has yet been found that would be

suitable for routine clinical practice in RCC (Nogueira & Kim 2008, Volpe &

Patard 2010).

Markers of the hypoxia-inducible pathway

The prognostic impact of markers of the hypoxia-inducible pathway is

controversial, indicating that the signalling network may be more complicated

than previously thought. Somatic VHL alterations are associated with a

favourable prognosis (Yao et al. 2002). In clear cell RCC, cytoplasmic HIF-1α

expression correlates with a more favourable prognosis, whereas nuclear

immunoexpression of HIF-1α is related to poor prognosis in metastatic disease

(Lidgren et al. 2005, Klatte et al. 2007). It has been suggested that HIF-2α might

be more important in the tumorigenesis of clear cell RCC than HIF-1α since with

dysfunctional VHL the transcription of hypoxia-regulated genes such as VEGF

33

has been shown to be regulated by HIF-2α (Nogueira & Kim 2008). However,

there are no reports of the prognostic value of HIF-2α immunoexpression as yet.

The overexpression of downstream targets of VHL and HIF-1α are associated

with both more and less favourable survival as carbonic anhydrase IX (CAIX)

was shown to be a marker of better prognosis whereas VEGF, on the other hand,

is related to worse survival rates (Bui et al. 2003, Bui et al. 2004, Jacobsen et al. 2004).

Markers of the mammalian target of rapamycin pathway

A loss of the tumour suppressor gene phosphatase and tensin homologue (PTEN)

activity and the following activation of the mTOR pathway are related to

decreased survival in RCC (Pantuck et al. 2007).

Markers of cell proliferation, cell cycle regulation and apoptosis

Cell proliferation, regulation of the cell cycle and programmed cell death

(apoptosis) are key factors in the initiation and progression of malignant tumours

(Linden et al. 1992, Hofstädter et al. 1995). However, it has been suggested that

in RCC mitotic activity is too low to provide survival information in individual

cases, but it could be associated with survival rates in larger populations.

Proliferation markers such as proliferating cell nuclear antigen (PCNA, cyclin),

Ki-67, argyrophilic nucleolar organizer regions (AgNOR) and spermine (SAT1)

have been reported to be correlated with prognosis. (Delahunt et al. 1993, Bui et al. 2004, Kallio et al. 2004, Pantuck et al. 2007, Nogueira & Kim 2008) Cell

cycle regulators p53 and p27 (cyclin-dependent kinase inhibitor 1B, CDKN1B,

Kip1) are independent predictors of poor survival in clear cell RCC (Migita et al. 2002, Cho et al. 2005, Shvarts et al. 2005, Nogueira & Kim 2008).

The prognostic role of apoptosis-related genes and proteins has not been

throroughly elucidated and mixed observations have been reported regarding their

prognostic value in RCC (Lam et al. 2008). The B-cell lymphoma 2 (Bcl-2)

family of apoptosis-regulating proteins, including Bcl-2 and the Bcl-2-associated

factor X (Bax), have been correlated with patient survival (Kallio et al. 2004). In

addition, the expression of p21 (cyclin-dependent kinase inhibitor 1, CDK-

interacting protein, WAF1, cip1), survivin and DIABLO (second mitochondria-

derived activator of caspaces, SMAC) have been associated with patient outcome

in RCC (Mizutani et al. 2005, Parker et al. 2006, Weiss et al. 2007).

34

Markers of cell adhesion

A loss of cellular adhesion between neoplastic cells and between neoplastic cells

and the extracellular matrix is one of the key processes in the development of

metastatic disease (Freemont & Hoyland 1996, van Kilsdonk et al. 2010). Cell

adhesion molecules have been studied as potential prognostic biomarkers for

RCC with variable results. The down-regulation of E-cadherin and cadherin-6 is

correlated with a poor prognosis in RCC but the prognostic significance of N-

cadherin seems to be quite the opposite (see section 2.6.2) (Katagiri et al. 1995,

Shimazui et al. 2006). The prognostic value of the cell-surface glycoprotein

CD44 seems to be debatable (Méjean et al. 2003). Expression of the epithelial

cell adhesion molecule (EpCAM) and the vascular cell adhesion molecule-1

(VCAM-1) have been related to a more favourable prognosis in RCC, whereas

the L1 cell adhesion molecule (L1CAM) and cellular adhesion molecule EphA2

have been shown to be associated with advanced disease and decreased survival,

respectively (Herrem et al. 2005, Nogueira & Kim 2008, Eichelberg et al. 2009).

Markers of degradation of the extracellular matrix

Matrix metalloproteinases (MMPs) degrade components of the extracellular

matrix and basement membrane, allowing tumour to grow, invade and metastasize.

Increased expression of MMPs has been demonstrated to be associated with

impaired prognosis in RCC. (Kallakury et al. 2001, Nogueira & Kim 2008) Also,

the expression of urokinase-type plasminogen activator (uPA), its receptor uPAR

and plasminogen activator inhibitor type 1 (PAI-1), which are involved in

regulating the degradation of the extracellular matrix, have been related to the

outcome of RCC patients (Nogueira & Kim 2008).

Markers of immune protection

It has been established that renal cell carcinoma induces the apoptosis of T-cells

and thus promotes tumorigenesis by suppressing host antitumour immunity

(Rayman et al. 2004). The B7 family is a group of co-regulatory ligands that

regulate T-cell-mediated immunity: B7-H1 and B7-H4 suppress immune

responses and their increased expression correlates with an unfavourable patient

outcome (Nogueira & Kim 2008). A cytokine receptor CXCR3 regulates tumour-

mediated immunity, angiogenesis and metastatic spread, and it is an independent

35

marker of a more favourable prognosis in RCC (Klatte et al. 2008). On the other

hand, the overexpression of another chemokine receptor, CXCR4, is related to

poor RCC-specific survival (Eichelberg et al. 2009). The soluble immunological

factors interleukin-6 (IL-6), interleukin-12 (IL-12) and the interleukin-2 receptor

(IL-2R) have also been correlated with the prognosis of RCC patients (Kallio et al. 2001).

Other prognostic biomarkers

Of the numerous tissue markers studied, for example insulin-like growth factor II

mRNA-binding protein IMP3 and vimentin are associated with an unfavourable

prognosis, whereas fascin has only been correlated with negative

clinicopathological features of tumours (Thompson et al. 2006, Eichelberg et al. 2009).

Some of the markers studied, such as CAIX, VEGF, amyloid A, insulin-like

growth factor 1 (IGF-1) and NSE, as well as some immunological factors, have

been demonstrated to predict patient survival when analysed in patient blood or

serum samples (Takashi et al. 1989, Rasmuson et al. 1993, Kallio et al. 2001,

Eichelberg et al. 2009). The presence of circulating tumour cells, endothelial cells

and tumour enzymes also has some prognostic value in RCC (Bluemke et al. 2009, Namdarian et al. 2010, Nisman et al. 2010).

The prognostic value of the quantification of tumour vasculature

(microvascular density) is controversial (Lam et al. 2008). The cytogenetic profile

of the tumour is correlated with the prognosis. In addition, many signature genes

have been recognized that, when used as a gene pattern, predict the clinical

outcome of patients (Nogueira & Kim 2008, Klatte et al. 2009a).

2.4.5 Prognostic factors in metastasized in renal cell carcinoma

In advanced RCC, classic anatomical and histological features of the primary

tumour have traditionally had limited predictive value (Volpe & Patard 2010).

The prognostic factors that have been identified in metastasized disease are

performance status, number and locations of metastatic sites, time to appearance

of metastases, prior nephrectomy and curative surgical resection of metastases

(Méjean et al. 2003). Bone metastases have been traditionally regarded as a

marker of shorter survival. However, it has been shown that the number of

metastatic sites is a more important prognostic marker than the location. (Lam et

36

al. 2008) Some laboratory findings such as low haemoglobin and elevated lactate

dehydrogenase, corrected serum calcium and inflammatory markers have been

correlated with patient survival (Volpe & Patard 2010).

2.4.6 Prognostic systems and nomograms

In order to improve predictive accuracy and categorize patients into different risk

groups, several algorithms called nomograms have been developed by combining

previously known prognostic factors. The most frequently used prognostic models

in localized RCC are the University of California Los Angeles Integrated Staging

System (UISS), which includes both clinical and pathological parameters, and the

Mayo Clinic’s Stage, Size, Grade and Necrosis (SSIGN) score. The accuracy of

both of these nomograms has been shown to be approximately 80%, which is

superior to TNM staging alone. (Zisman et al. 2001, Frank et al. 2002, Volpe &

Patard 2010) The Kattan nomogram was designed to estimate the probability of 5-

year recurrence-free survival after nephrectomy and after its update in 2005 its

accuracy is about 80% as well (Kattan et al. 2001, Volpe & Patard 2010). In a

novel prognostic model by Karakiewicz et al. both the clinical and pathological

variables combined within the algoritm were shown to provide a predictive

accuracy of over 80% (Karakiewicz et al. 2007).

The most widely used prognostic tools for advanced RCC are the Memorial

Sloan-Kettering Cancer Center (MSKCC) and the French group of

immunotherapy instruments, which combine clinical parameters including

performance status, serum laboratory examinations, time from diagnosis to

iniatiation of systemic therapy and also the number and location of metastatic

sites in order to stratify patients into different risk categories. Lately, new

prognostic models have been developed and old ones updated in order to predict

patient outcomes under molecular targeted therapy. (Motzer et al. 1999, Volpe &

Patard 2010)

The addition of set of molecular markers to the nomogram of classic

variables could improve the predictive accuracy considerably (Kim et al. 2004,

Klatte et al. 2009b). Prognostic models that combine various independent

prognostic molecular markers have been developed, such as the biomarker panel

BioScore that includes B7-H1, survivin and Ki-67 (Parker et al. 2009). The

molecular signature, which consists of Ki-67, p53, VEGF and VEGF receptors

(VEGFR), is more accurate than clinical or pathological prognostic factors alone

in predicting patient outcome in RCC (Klatte et al. 2009b).

37

2.5 Treatment

2.5.1 Treatment of localized renal cell carcinoma

Historially, the gold standard curative therapy for patients with localized RCC has

been radical nephrectomy (Robson et al. 1969). Nowadays, partial nephrectomy

(nephron-sparing surgery, NSS) is a recommended standard treatment for local

renal tumours up to a diameter of 7 cm and an option for larger, local tumours

whenever technically feasible (Uzzo & Novick 2001, Ljungberg et al. 2010, Van

Poppel et al. 2011). The oncological efficacy of NSS has been confirmed in large,

prospective randomized studies (Van Poppel et al. 2011), and there is evidence to

show that NSS decreases the risk of renal failure and relative adverse effects on

general health, including cardiovascular events, compared to radical nephrectomy

(Huang et al. 2009).

With its adherence to the principles of open radical surgery, laparoscopic

renal surgery is considered a standard of care for selected RCC patients and it is

associated with a lower rate of morbidity than open surgery (Link et al. 2005,

Ljungberg et al. 2010). Laparoscopic nephrectomy is a recommended standard

care for local tumours that are not suitable for NSS and it provides equivalent

oncologic outcomes to open surgery (Patard et al. 2004, Becker et al. 2006,

Ljungberg et al. 2010). In experienced hands and with carefully selected patients,

laparoscopic NSS is an alternative to open partial nephrectomy. Robotic-assisted

partial nephrectomy is under evaluation. (Ljungberg et al. 2010)

In surgical treatment of RCC the official role of lymphadenectomy is

restricted to staging purposes, principally at the hilar region. In selected cases

with lymph node metastases restricted to the retroperitoneal area, extended

lymphadenectomy may improve RCC patient survival rates. When the adrenal

gland is normal in preoperative imaging, routine adrenalectomy is only

recommended for large upper pole tumours or tumours with a diameter greater

than 7 cm. (Ljungberg et al. 2010)

Minimally-invasive techniques, for example, percutaneous radiofrequency

(RF) ablation (Lui et al. 2003), cryoablation (Gill et al. 2005), microwave

ablation (Johnson & Nakada 2001), laser ablation (Hinshaw & Lee 2004) and

high-intensity focused ultrasound (HIFU) (Janzen et al. 2002) are alternatives for

surgical treatment in selected RCC patients with decreased performance status or

multiple tumours. In addition, small renal tumours may be actively followed and

treated if they progress. (Ljungberg et al. 2010)

38

2.5.2 Treatment of metastasized renal cell carcinoma

Resistance against radiotherapy, cytotoxic drugs and hormones is a typical feature

of RCC (DiBiase et al. 1997, Finley et al. 2011). Immunotherapy with interferon

alpha (IFN-α) or interleukin-2 (IL-2) can produce complete and durable responses,

although response rates have generally been shown to be 6–15% for IFN-α and 7–

27% for IL-2, providing only a modest survival benefit in advanced RCC.

Nowadays, the official role of immunotherapy has been restricted to adjuvant use

with bevacizumab. (Ljungberg et al. 2010, Finley et al. 2011)

Molecular targeted therapies are designed to block the critical signalling

pathways underlying the pathogenesis of RCC. They are categorized into three

groups: tyrosine kinase and multikinase inhibitors, VEGF antibodies and mTOR

inhibitors. (Hutson 2011) Targeting agents are a recommended systemic treatment

for metastasized RCC (Ljungberg et al. 2010). The efficacy of molecular targeted

therapies has been proven in clinical trials where they have improved both

progression-free and overall survival. Durable remissions are, however, rare

because these drugs do not eradicate the disease. (Finley et al. 2011, Hutson 2011)

The advantages of molecular targeted therapies compared to immunotherapy are

better efficacy and tolerability, as well as oral administration (Hutson 2011).

Cytoreductive nephrectomy and resection of metastases improve patient

survival and are recommended for patients with a good performance status (van

der Poel et al. 1999, Flanigan et al. 2004, de Reijke et al. 2009, Ljungberg et al. 2010). In the era of molecular targeted therapies the role and timing of palliative

surgery have, however, been blurred, and they should be re-evaluated in clinical

trials (Polcari et al. 2009).

Palliative radiotherapy is indicated for symptomatic brain and bone

metastases that do not respond to systemic treatment. The embolization of bone

and paravertebral metastases can also relieve pain caused by metastases.

(Ljungberg et al. 2010)

2.6 Reasoning for studied biomarkers

2.6.1 Oxidative stress in carcinogenesis and markers of the

oxidative system

Oxidative stress is a state where the amount of reactive oxygen species (ROS)

exceeds the antioxidant capability of the target cell (Chance et al. 1979, Klaunig

39

et al. 1998). In addition to mitochondrial oxidative phosphorylation and some

other endogenous sources of ROS, ionizing radiation, ozone, various drugs,

hormones and other xenobiotic chemicals alter the oxidative status in cells

(reviewed in Klaunig et al. 1998, Oberley 2002). At low levels, ROS have

physiological functions such as the regulation of signal transduction pathways,

transcription factors and mitochondrial enzyme activity. A moderate increase in

ROS activity promotes cell proliferation and differentiation, but excessive

oxidative stress is harmful for cells. (reviewed in Oberley 2002, Trachootham et al. 2009)

Oxidative stress promotes carcinogenesis via several mechanisms that cover

cancer initiation, promotion and progression. Reactive oxygen species cause

structural alterations in DNA, prevent DNA damage repair and activate chemical

carcinogens. (reviewed in Klaunig et al. 1998) In addition, ROS activate proto-

oncogenes, inactivate tumour suppressor genes and modify cytoplasmic and

nuclear signal transduction. By activating transcription factors, ROS are involved

in gene expression regulation. (reviewed in Toyokuni et al. 1995, Wiseman &

Halliwell 1996, Klaunig et al. 1998) Oxidative stress-mediated signalling events

have been shown to contribute to various functions involved in carcinogenesis,

such as cell proliferation and survival, apoptosis, energy metabolism, cell

morphology, cell-cell adhesion, cell motility and metastasis, angiogenesis and the

maintenance of tumour stemness (reviewed in Liou & Storz 2010). Finally,

oxidative stress induces chemoresistance against anticancer drugs by activating

specific antioxidant systems (reviewed in Toyokuni et al. 1995).

Persistent oxidative stress is present in cancer cells, including RCC

(Szatrowski & Nathan 1991, Okamoto et al. 1994, Toyokuni et al. 1995).

Simultaneously with increased ROS production there may also be an increase or

decrease in antioxidants and ROS-scavenging enzymes, which makes regulation

of the redox balance quite complex in cancer cells (reviewed in Trachootham et al. 2009). High levels of oxidative stress have been shown to be associated with

aggressive behaviour and poor prognosis in cancer (reviewed in Kumar et al. 2008, Trachootham et al. 2009). In RCC, oxidative DNA damage has been shown

to be related to poor differentiation (Soini et al. 2006b).

Nfr2 and Keap1

The erythroid transcription factor NF-E2 (nuclear factor erythroid-2-related factor

2, NF-E2-related factor 2, Nfr2) is a nuclear transcription factor that regulates

40

genes encoding antioxidants, xenobiotic detoxifying enzymes and drug efflux

pumps in response to oxidative, electrophilic and xenobiotic stresses (Itoh et al. 1997). It is negatively regulated by Kelch-like ECH-associated protein 1 (BTB-

Kelch-type substrate adaptor protein, Keap1, INrf2) (Itoh et al. 1999). The Keap1

protein is a cysteine-rich cytoplasmic protein that binds to actin. It is localized in

the nuclear periphery at the cytoskeleton and in adhesion complexes. (Itoh et al. 2004) The primary task of Keap1 is to sense xenobiotic and oxidative stresses and

to modulate Nfr2 stability (Motohashi & Yamamoto 2004). Under normal

conditions, the Neh2 domain of Nfr2 is bound to the double glycine repeat (DGR)

domain of Keap1, which results in constant ubiquitination and proteosomal

degradation of Nfr2 and further suppression of Nfr2 activity. In addition, Keap1

regulates Nfr2 by altering its subcellular location. (McMahon et al. 2003, Itoh et al. 2004) Oxidative stress antagonizes the interaction between Nfr2 and Keap1 by

modifying cysteine residues of Keap1, resulting in the dissociation of Nfr2 from

Keap1 and the accumulation of Nfr2 (McMahon et al. 2003, Yamamoto et al. 2008). When the Keap1-Nfr2 complex disrupts, Nfr2 is able to translocate into

the nucleus where it binds to the antioxidant response element (ARE) and

activates the transcription of various detoxification and antioxidant proteins. Then,

Nfr2 activation is switched off and Nfr2 is exported back to the cytoplasm where

it binds to Keap1 and is degraded. (Kaspar et al. 2009) The Nfr2-Keap1 system is

regarded as being one of the most important cellular defence mechanisms against

oxidative and xenobiotic stresses (Motohashi & Yamamoto 2004).

It is assumed that either too much or too little Nfr2 activity can result in

physiological impairments (Liu et al. 2005). The biological consequences of Nfr2

activation are ambiguous in carcinogenesis. On the one hand, Nfr2 activity

protects DNA against oxidative and xenobiotic damage by eliminating ROS and

detoxifying carcinogens and, thus, it may suppress cancer initiation. It may also

prevent metastasis by affecting the redox balance. (Satoh et al. 2010, Taguchi et al. 2011) On the other hand, Nfr2 activation can promote tumourigenicity and

metastasis via several ways, such as by facilitating proliferation, anchorage-

independent growth and cell survival, as well as by inhibiting apoptosis (Singh et al. 2008, Satoh et al. 2010, Shibata et al. 2010, Taguchi et al. 2011). Moreover,

Nfr2 mutation activates the mTOR pathway (Shibata et al. 2010). Through the

up-regulation of antioxidant enzymes and drug efflux pathways, Nfr2 activation

induces chemoresistance (Singh et al. 2008).

Mutations in the Keap1 or Nfr2 genes result in impairments in Keap-

mediated Nfr2 regulation and increased Nfr2 activity. Gain-of-function mutations

41

of the Nfr2 gene have mainly been reported in squamous cell carcinomas such as

head and neck, lung, oesophageal and skin carcinomas and result in inactivated

Keap1-mediated degradation of Nfr2 and the nuclear accumulation of Nfr2

(Shibata et al. 2008b, Kim et al. 2010). Loss-of-function mutations in the Keap1

gene have been described in several adenocarcinomas, including breast,

gallbladder and lung cancers (Singh et al. 2006, Nioi & Nguyen 2007, Shibata et al. 2008a). In lung cancer, Keap1 mutations are associated with decreased Keap1

immunoexpression (Ohta et al. 2008). In head and neck squamous cell carcinoma,

increased Nfr2 expression has been reported in 92% of tumours at the same time

as increased expression of the Nfr2-downstream target thioredoxin and the

relative of overexpression Keap1 (Stacy et al. 2006).

Mutations of the Nfr2 and Keap1 genes have been found to be associated

with poor survival in both lung and head and neck carcinomas (Shibata et al. 2010, Takahashi et al. 2010). Decreased Keap1 expression and the resulting

nuclear accumulation of Nfr2 has been shown to stimulate tumour growth in lung

adenocarcinoma (Ohta et al. 2008) and to be correlated with worse overall

survival rates in non-small cell lung cancer (Solis et al. 2010).

Thioredoxin

Thioredoxin (Trx, adult T-cell leukaemia-derived factor, ADF) is a thiol-

containing antioxidant enzyme, the expression of which is regulated by Nfr2

(Laurent et al. 1964, Powis et al. 2000, Stacy et al. 2006). Thioredoxin is

predominantly a cytosolic protein but it has also been detected in the nucleus in

malignant cells. The expression of Trx is cell-cycle dependent. Hypoxia and

oxidative stress are able to induce Trx expression as well as facilitate its

translocation into the nucleus. The antioxidant function of Trx is primary based

on its ability to donate electrons for oxidized thioredoxin peroxidase, which then

removes oxidant species. The catalytic site (-Trp-Cys-Gly-Pro-Cys-Lys-) of Trx is

reversibly oxidized to the cysteine disulphide Trx-S2. (reviewed in Powis et al. 2000, Powis & Montfort 2001) In addition, Trx acts directly by removing

hydrogen peroxide (Spector et al. 1988).

Thioredoxin may promote tumorigenesis via several mechanisms. Trx

stimulates tumour cell growth by sensitizing tumour cells to growth factors

produced by the tumour itself (Gasdaska et al. 1995). It also regulates

transcription factors involved in tumorigenesis, such as nuclear factor-kappaB

(NF-kappaB), activator protein 1 (AP-1) and HIF-1α, as well as the tumour

42

suppressor gene p53 (reviewed in Powis & Montfort 2001), and it inhibitis

apoptosis and thus offers both survival and growth advantages for cancer cells

(Baker et al. 1997). The association between Trx expression and cell proliferation

is contradictory. In some cancers, Trx expression is positively associated with cell

proliferation activity (Grogan et al. 2000, Turunen et al. 2004), but the opposite

situation has also been reported (Rubartelli et al. 1995). In addition, Trx facilitates

chemoresistance (Singh et al. 2008).

Secretion of Trx has been shown in neoplastic cells (Rubartelli et al. 1992).

Increased expression of Trx has been observed in several malignomas, for

example in cervix, colorectal, gastric and non-small cell lung carcinomas (Fujii et al. 1991, Gasdaska et al. 1994, Berggren et al. 1996, Grogan et al. 2000).

Thioredoxin overexpression is associated with features of more aggressive

tumour growth, such as increased proliferation and decreased apoptotic activity

(reviewed in Powis & Montfort 2001). In colorectal carcinoma, Trx is an

independent marker of shortened survival rates (Raffel et al. 2003). In RCC, the

immunoexpression of Trx is associated with older age and a lower apoptotic index

(Soini et al. 2006b).

2.6.2 Cell adhesion, migration, invasion and metastasis and related proteins

The ability to invade other tissues and spread to distant organs is chraracteristic of

cancer cells (Hanahan & Weinberg 2011). Tumour metastasis is the most

important cause of mortality in cancer. It is a complex, multistep process that

begins when tumour cells lose their cell-cell attachments, dissociate from the

primary tumour and invade the extracellular matrix through the epithelial

basement membrane. After entering the blood stream or lymphatic vessels,

tumour cells disseminate via the circulation to distant organs where they form

secondary lesions. The key processes in metastasis are changes in cellular

adhesion, the production of proteolytic enzymes, which degrade the extracellular

matrix, and the secretion of various cytokines, which promote invasion and

angiogenesis. (reviewed in Meyer & Hart 1998)

Cell adhesion molecules are transmembrane glycoproteins that mediate

interactions between cells and between cells and the tissue matrix. Changes in cell

adhesion allow tumour cells to migrate, invade and metastasize. In addition, cell

adhesion molecules are involved in cell signalling, proliferation and tumour

growth. (reviewed in Freemont & Hoyland 1996, van Kilsdonk et al. 2010) Cell

43

adhesion molecules are divided into four major groups; one of these is the large

cadherin family which mediates calcium-dependent intercellular adhesion. They

interact with the cytoskeleton through cytoplasmic regulatory proteins called

catenins. (Takeichi et al. 1988, reviewed in Makrilia et al. 2009)

E-cadherin

Epithelial cadherin (E-cadherin, uvomorulin, L-CAM, cell-CAM 120/80, CD324)

(Sommers et al. 1991, van Kilsdonk et al. 2010) mediates homotypic and

homophilic cell-cell adhesion in the epithelium and is essential for epithelial

integrity and polarity. It consists of an extracellular domain, a transmembrane

segment and a cytoplasmic domain. (reviewed in Freemont & Hoyland 1996,

Chetty & Serra 2008, Makrilia et al. 2009) The cytoplasmic domain is complexed

with either beta- or gamma-catenin (plakoglobin), which then binds directly to

alpha-catenin and thus connects the E-cadherin-catenin complex to the actin

cytoskeleton. (Aberle et al. 1996)

The loss of E-cadherin promotes tumorigenesis via several routes. The

decreased expression of E-cadherin disrupts cell adhesion, allowing tumour cells

to detach, and it facilitates conversion to a more migratory phenotype and

invasion and metastasis (reviewed in van Kilsdonk et al. 2010). The loss of E-

cadherin is a crucial step in the epithelial to mesenchymal transition (EMT),

which is related to a tumour’s ability to metastasize (see section 2.3.4) (reviewed

in Banumathy & Cairns 2010). In addition, the loss of E-cadherin liberates beta-

catenin, which translocates to the nucleus and activates the transcription of genes

responsible for cell proliferation, differentiation and migration (see section 2.3.3)

(reviewed in Banumathy & Cairns 2010, van Kilsdonk et al. 2010). Several

genetic and epigenetic mechanisms have been found that inactivate E-cadherin-

mediated cell adhesion and promote tumour invasiveness and progression. The

mechanisms that result in impaired E-cadherin functioning are reported to be

DNA hypermethylation around the promoter region of the E-cadherin gene,

mutations in E-cadherin and catenin genes, the transcriptional inactivation of E-

cadherin, tyrosine phosporylation of the E-cadherin/beta-catenin complex and the

phosphorylation of beta-catenin. In addition, several proteases such as some

matrix metalloproteases are capable of cleaving E-cadherin; the soluble E-

cadherin fragments released inhibit E-cadherin functioning in a paracrine way. A

cell membrane glycoprotein called dysadherin also down-regulates E-cadherin.

(reviewed in Makrilia et al. 2009)

44

The disruption of E-cadherin-mediated cell-cell adhesion followed by a more

invasive and aggressive phenotype has been established in many cancers of

epithelial origin, such as squamous cell carcinomas of the head and neck and

oesophagus and carcinomas of the bladder, breast, cervix, colon, endometrium,

lung, pancreas, prostate and thyroid. In addition, decreased E-cadherin expression

has been shown in precancerous conditions such as Barrett’s oesophagus and in

the progression of dysplasia to adenocarcinoma. (Ilyas & Tomlinson 1997;

reviewed in Chetty & Serra 2008, Makrilia et al. 2009) E-cadherin negativity is

characteristic of invasive lobular breast carcinoma and diffuse-type gastric cancer

(Gamallo et al. 1993, Becker et al. 1999).

In general, the loss of E-cadherin is associated with decreased differentiation,

invasiveness and metastatic potential and higher mortality rates in cancer (Ilyas &

Tomlinson 1997, reviewed in Makrilia et al. 2009). The relationship between

decreased E-cadherin immunoexpression and unfavourable clinicopatohological

tumour features was previously established, for example in breast, colon and

oesophageal carcinomas (Sommers et al. 1991, Doki et al. 1993, Dorudi et al. 1993). For example in bladder, prostate and oesophageal carcinomas decreased E-

cadherin immunoepxression is a marker of poor patient prognosis (Giroldi et al. 1994, Tamura et al. 1996). However, the prognostic significance of the loss of E-

cadherin is not independent of tumour stage and grade (Ilyas & Tomlinson 1997).

Inactivation of the VHL-HIF pathway is a critical step in the development of

clear cell RCC (Gnarra et al. 1994, Kaelin 2004) and a direct association between

VHL status and E-cadherin expression has been established in clear cell

carcinoma (Esteban et al. 2006, Krishnamachary et al. 2006, Evans et al. 2007). It

is believed that the loss of E-cadherin is an early pathogenic event in clear cell

RCC (Esteban et al. 2006, Russell & Ohh 2007). The loss of VHL and the

subsequent accumulation of HIF-1 reduce E-cadherin expression via the action of

various transcriptional repressors (Krishnamachary et al. 2006, Evans et al. 2007).

Hypermethylation of the E-cadherin gene was reported in 11% of primary RCCs

(Dulaimi et al. 2004). The decreased expression of E-cadherin was established in

several studies on RCC (Terpe et al. 1993, Katagiri et al. 1995, Shimazui et al. 1997, Kuroiwa et al. 2001) and the loss of E-cadherin immunoexpression was

associated with advanced stage and decreased survival (Katagiri et al. 1995).

45

Beta-catenin

Beta-catenin (β-catenin) is a member of the regulatory cytoplasmic protein family

called catenins and is a vertebrate homologue of the Armadillo protein in

Drosophila. It is directly associated with the cytoplasmic domain of E-cadherin

and forms a structural link between E-cadherin and the actin cytoskeleton via

alpha-catenin. In addition to its essential role in cell adhesion as a component of

the E-cadherin complex, beta-catenin is involved in intracellular signal

transduction. (reviewed in Gumbiner 1995) It is believed that beta-catenin could

be a connection between cell adhesion and mitogenic signalling (reviewed in

Karim et al. 2004).

Beta-catenin is a part of the Wnt pathway (see section 2.3.3) and the

accumulation of beta-catenin seems to play a central role in oncogenesis. Beta-

catenin promotes the transcription of target genes and is involved in the regulation

of cell proliferation, differentiation and migration, as well as the inhibition of

apoptosis. (Behrens et al. 1996, reviewed in Karim et al. 2004) On the other hand,

a lack of beta-catenin disrupts E-cadherin-mediated cell adhesion, which has also

been implicated in carcinogenesis (reviewed in Makrilia et al. 2009). It is

believed that the overexpression of beta-catenin is important in tumour initiation,

whereas the loss of beta-catenin and the resulting loss of function of the E-

cadherin-beta-catenin complex and impaired cellular adhesion are involved in

tumour invasion. The cytoplasmic level of beta-catenin is regulated by Wnt and

APC, where Wnt increases the expression of beta-catenin and, in turn, APC

decreases it. (Ilyas & Tomlinson 1997) In addition, as E-cadherin binds beta-

catenin it recruits it to the cell membrane and prevents its nuclear translocation

(Orsulic et al. 1999).

Beta-catenin gene mutations and/or accumulation have been found in several

malignancies such as colorectal, gastric, hepatocellular, prostate, oesophageal,

and anaplastic thyroid carcinomas, as well as lung adenocarcinoma and malignant

melanoma (reviewed in Karim et al. 2004). On the other hand, decreased levels of

beta-catenin expression have also been reported in cancers such as colon, gastric

and oesophageal carcinomas (reviewed in Takayama et al. 1996, Ilyas &

Tomlinson 1997). The prognostic significance of beta-catenin is not similar in

different cancers. For example, in colon and liver cancers increased levels of

cytoplasmic and nuclear beta-catenin are associated with invasion and poor

prognosis, whereas in melanoma they are related to a favourable outcome

(Arozarena et al. 2011).

46

The involvement of the Wnt/beta-catenin pathway has been established in the

pathogenesis of RCC and beta-catenin is regarded as one of the key molecules in

the oncogenesis of RCC. In animal studies, the overexpression of beta-catenin

was shown to induce renal tumours. (reviewed in Banumathy & Cairns 2010)

Mutations in the beta-catenin gene are, however, rare in RCC and are only

reported in clear cell variants. Instead, the accumulation of beta-catenin is a more

frequent event which suggests that there are other mechanisms activating the Wnt

signalling pathway in RCC. (Kim et al. 2000) For example, hypermethylation of

the APC gene promoter, splicing isoforms of the transcriptional factor TCF-4 and

the loss of various Wnt antagonists may regulate beta-catenin expression and Wnt

signalling (Battagli et al. 2003, Shiina et al. 2003, Awakura et al. 2008, Hirata et al. 2009, Kawakami et al. 2009, Kojima et al. 2009). There are no APC tumour

suppressor gene mutations in RCC (Kim et al. 2000). It has been found that VHL

down-regulates beta-catenin by stabilizing a protein named Jade-1 (Chitalia et al. 2008). In addition, VHL represses the HGF-driven oncogenic signalling of beta-

catenin (see section 2.3.5) (Peruzzi et al. 2006).

The immunoexpression of beta-catenin has previously been reported in RCC

(Shimazui et al. 1997, Bilim et al. 2000, Kim et al. 2000, Zhu et al. 2000, Guo et al. 2001, Kuroiwa et al. 2001, Aaltomaa et al. 2004). Previous studies also

showed the prognostic significance of beta-catenin expression in RCC (Shimazui et al. 1997, Bilim et al. 2000, Aaltomaa et al. 2004).

Myosin VI

Myosins are a large family of molecular motor proteins that hydrolyse adenosine-

5'-triphosphate (ATP) and convert chemical energy into a mechanical force in

order to move along actin filaments. Myosin VI is one of the so-called

unconventional myosins and was first identified in Drosophila melanogaster. It

moves in the reverse direction compared to other myosins, i.e. towards the

pointed (-) end of actin. (Kellerman & Miller 1992, Wells et al. 1999) The

movement of myosin VI is directed away from the plasma membrane into the cell

and away from the surfaces of internal organelles such as the Golgi complex

(Buss et al. 2001, Buss et al. 2004). Intracellularly, myosin VI is located in

clathrin-coated endocytic vesicles, membrane ruffles and the Golgi complex

(Buss et al. 2001, Roberts et al. 2004), but nuclear myosin VI expression has also

been described (Vreugde et al. 2006). Myosin VI forms a complex with E-

cadherin and beta-catenin and protects them from degradation. Similarly, E-

47

cadherin and beta-catenin preserve myosin VI expression. (Küssel-Andermann et al. 2000, Geisbrecht & Montell 2002)

Myosin VI has multiple roles that are connected to oncogenesis (Knudsen

2006, Sweeney & Houdusse 2007). By anchoring to the actin cytoskeleton,

myosin VI maintains cell shape and rigidity (Millo et al. 2004). Myosin VI

strengthens cellular adhesion as it provides tension between the E-cadherin-beta-

catenin cell adhesion complex and the cytoskeleton. In addition, myosin VI fixes

epithelial cells to the basement membrane and thus maintains the polarity of

epithelial cell layers. (Küssel-Andermann et al. 2000, Geisbrecht & Montell 2002,

Millo et al. 2004) By connecting to membrane proteins such as E-cadherin,

myosin is able to produce the protrusion force needed in cell migration and is thus

involved in cell motility (Geisbrecht & Montell 2002). Myosin VI has several

transport roles, such as endocytosis and secretion (Buss et al. 2001, Warner et al. 2003); it carries cargo from the periphery towards the centre of the cell via

clathrin-mediated endocytosis (Buss et al. 2001). Myosin VI-mediated membrane

transport has a crucial role in cell division and the inhibition of myosin VI

expression results in a defect in cytokinesis (Arden et al. 2007). In addition,

myosin VI modulates gene expression and receptor clustering (Dunn et al. 2006,

Vreugde et al. 2006, Sweeney & Houdusse 2007), and it may form a link between

endocytosis and signalling pathways (Buss et al. 2001).

The overexpression of myosin VI has been demonstrated in human cancers

such as ovarian and prostate carcinomas (Yoshida et al. 2004, Dunn et al. 2006,

Loikkanen et al. 2009). Myosin VI expression is related to malignant properties

such as cancer cell migration and dissemination, as well as to the aggressive

clinical behaviour of tumours (Geisbrecht & Montell 2002, Yoshida et al. 2004,

Dunn et al. 2006). In the normal kidney, myosin VI is highly expressed at the

base of the brush border in proximal tubule cells (Biemesderfer et al. 2002).

2.6.3 Inflammation-related cancer and the immunologic nature of renal cell carcinoma

The immune system consists of innate and acquired (adaptive) immunity that

protect the host from various foreign substances and damaging agents such as

microbes and cancer cells (reviewed in Reddy & Reddy 2010). The first line of

defence against infections is based on innate immunity, whereas the initiation of

acquired immunity takes several days. Innate immunity is composed of

phagocytic cells such as macrophages, neutrophils and dendritic cells, which

48

digest pathogens in a non-specific manner and present pathogen-derived antigens

to T-lymphocytes. Acquired immunity consists of T- and B-lymphocytes, which

are capable of recognizing specific peptide antigens through antigen receptors

located on their surfaces. Acquired immunity is characterized by specificity and

memory. However, innate immunity is also capable of recognizing pathogen-

associated molecular patterns through Toll-like receptors (TLRs) on cell surfaces.

(reviewed in Akira & Hemmi 2003, Wagner 2004) The B-cells are involved in the

humoral immune response, which is based on secreted antibodies. Antigen-

specific cytotoxic T-cells are capable of inducing apoptosis and account for cell-

mediated immunity along with macrophages and natural killer (NK) cells.

(reviewed in Reddy & Reddy 2010) In addition, cells of the immune system, like

helper T-cells, can secrete various immunomodulating agents such as interferons

(IFNs), interleukins (ILs) and tumour necrosis factor alpha (TNF-α), which may

be involved in several processes in the body (reviewed in Akira et al. 2001).

The relationship between inflammation and cancer development has been

recognized for years, but the exact mechanisms by which inflammation results in

carcinogenesis are not fully known. It is known that chronic inflammation causes

a repeated cycle of genomic damage and healing with accelerated cell

proliferation that promotes both malignant transformation and clonal expansion of

transformed cells. (reviewed in Farrow & Evers 2002) The peritumoural

inflammation observed could also indicate a relationship between renal

carcinogenesis and inflammation (Tuna et al. 2004).

Malignant tumours repress immunosurveillance as they produce

immunosuppressive mediators by themselves or they cause surrounding

inflammatory cells to release them (Huang et al. 1998). In RCC, the apoptosis of

T-cells is induced, which suppresses host antitumour immunity and thus promotes

tumorigenesis (Uzzo et al. 1999, Rayman et al. 2004). On the other hand,

inflammatory activation is associated with a more favourable clinical outcome in

some cancers such as melanoma (Hernberg et al. 1997). Bacterial DNA purified

from Bacillus Calmette-Guerin (BCG) stimulates NK cells and IFN production,

which have antitumour activity in cancer (Tokunaga et al. 1984).

Renal cell carcinoma is generally considered as one of the immunogenic

malignancies. Immunotherapy can produce significant responses in advanced

RCC (Vogelzang et al. 1992) and tumour vaccines have been shown to have some

benefits as well (Finley et al. 2011). Rare cases of the spontaneous regression of

RCC metastases are presumably caused by immunological mechanisms (Lokich

1997). Renal cell carcinomas attract several kinds of immune cells (Geiger et al.

49

2009) and various tumour-associated antigens have been detected that arouse

tumour-specific T-cell-defined immune responses (Neumann et al. 1998).

Immunological parameters such as the amount of peripheral blood and

intratumoural neutrophils and other immune cells, as well as serum cytokine

levels, have been shown to be related to the outcome of RCC patients (Kallio et al. 2001, Donskov & von der Maase 2006).

Cyclooxygenase-2

Cyclooxygenase (COX, prostaglandin endopedoxidase H2 synthase, PGHS) is a

key enzyme in the synthesis of prostaglandins (PGs) from arachidonic acid

(DeWitt & Smith 1988). The activity of COX can be inhibited by non-steroidal

anti-inflammatory drugs (NSAIDs) (Vane 1971). There are two isoforms of COX,

COX-1 and COX-2, which share a similar structure and catalytic activity but

differ in expression pattern and substrate and inhibitor selectivity (reviewed in

Smith et al. 1996, Vane et al. 1998). Cyclooxygenase-1 is permanently expressed

in almost all cell types; it catalyses the synthesis of protective PGs involved in

shielding gastric and intestinal mucosa and in maintaining normal renal function

in compromised kidneys. In platelets, the inhibition of COX-1 results in the loss

of aggregation. (reviewed in Vane et al. 1998)

Cyclooxygenase-2 expression is not observed in most mammalian tissues

under normal conditions. However, various stimuli including inflammatory

factors such as bacterial lipopolysaccharides and several cytokines, hormones,

growth factors, oncogenes and tumour promoters induce COX-2 expression

within hours. (Sheng et al. 1998; reviewed in Smith et al. 1996, Dubois et al. 1998, Vane et al. 1998) The induction of COX-2 can be decreased by anti-

inflammatory cytokines, corticosteroids and NSAIDs (reviewed in Vane et al. 1998). Cyclooxygenase-2 participates in numerous important physiological and

pathophysiological conditions such as inflammation, cell growth control, nerve

transmission (especially pain and fever), neural development and adaptation,

reproduction, bone metabolism and wound healing (reviewed in Dubois et al. 1998, Vane et al. 1998). The side effects of COX-2 inhibition include

cardiovascular toxity such as stroke, myocardial infarction and other

thromboembolic events (reviewed in Menter et al. 2010).

The relationship between COX-2 and carcinogenesis was originally observed

in epidemiological studies where aspirin use was associated with a reduced risk in

colorectal carcinoma (Thun et al. 1991). Up-regulation of COX-2 gene expression

50

was later established in colorectal carcinoma specimens (Eberhart et al. 1994).

There are several mechanisms via which COX-2 can promote carcinogenesis.

Indeed, COX-2 has also been shown to induce tumorigenic transformation in

transgenic mice (Liu et al. 2001). It catalyzes the conversion of pro-carcinogens

to proximate carcinogens (Eling et al. 1990) and enhances cell migration and

adhesion to the extracellular matrix, thus promoting invasion and metastasis

(Tsujii & DuBois 1995, Tsujii et al. 1997, Cao & Prescott 2002).

Cyclooxygenase-2 expression favors cancer cell growth by stimulating

proliferation (Sheng et al. 1998) and inducing tumour neoangiogenesis (Tsujii et al. 1998, Masferrer et al. 2000). Cyclooxygenase-2 overexpression also inhibits

apoptosis (Tsujii & DuBois 1995). It is assumed that COX-2 has an important role

in inflammation-induced carcinogenesis (Farrow & Evers 2002) and, in addition,

that it suppresses antitumour immunity (Huang et al. 1998, Williams et al. 1999).

The up-regulation of COX-2 has been reported in several malignomas,

including breast, colorectal, gastric and also renal cell carcinoma (Eberhart et al. 1994, Subbaramaiah et al. 1996, Ristimäki et al. 1997, Miyata et al. 2003). It is

associated with a poor clinical outcome in various cancers such as breast,

colorectal and oesophageal carcinomas (Sheehan et al. 1999, Buskens et al. 2002,

Ristimäki et al. 2002). In RCC, COX-2 is involved in angiogenesis as well as

invasiveness of the tumour (Chen et al. 2004a, Chen et al. 2004b).

HuR

The HuR protein (ELAVL1, HuA) is a member of the RNA binding protein

family ELAV (embryonic lethal abnormal vision), which is involved in diverse

physiological processes (Ma et al. 1996, reviewed in Hinman & Lou 2008). Gene

expression is regulated at the DNA and protein level and, in addition, it can be

altered via the modulation of messenger RNA (mRNA) stability and translation

(reviewed in Brennan & Steitz 2001). The HuR protein stabilizes mRNA by

binding to the adenylate and uridylate residues in the 3’ untranslated region (3’-

UTR), which are called AU-rich elements (AREs) (Ma et al. 1996, Fan & Steitz

1998). Furthermore, HuR is predominantly a nuclear protein that binds to ARE-

containing mRNAs in the nucleus and escorts them into the cytoplasm where it

protects them from degradation (Fan & Steitz 1998). Through its target mRNAs,

HuR has been linked to various physiological functions including cell growth,

differentiation and proliferation, as well as oncogenesis (Peng et al. 1998, López

de Silanes et al. 2003, reviewed in Brennan & Steitz 2001).

51

The HuR protein regulates COX-2 expression by stabilizing COX-2 mRNA

(Sengupta et al. 2003), which has been found in several types of carcinomas

including breast, colon, gastric and ovarian carcinomas (Dixon et al. 2001,

Erkinheimo et al. 2003, Denkert et al. 2004b, Mrena et al. 2005). In addition to

COX-2, HuR has other target mRNAs that are involved in carcinogenesis and, for

example, inhibit apoptosis and promote cell proliferation, invasion, metastasis and

angiogenesis (Levy et al. 1998, Wang et al. 2000, reviewed in López de Silanes et al. 2005). The HuR protein may also induce the expression of

immunosuppressive cytokines, which might help the tumour to avoid detection

via immune surveillance (Nabors et al. 2001, reviewed in López de Silanes et al. 2005).

In normal tissue, the expression pattern of HuR is weak or moderate and

primarily nuclear, whereas up-regulated and cytoplasmic HuR expression is

detected in virtually all cancer tissues, including RCC (López de Silanes et al. 2003). Mutations of the HuR gene have never been detected in cancer but there

are other mechanisms that increase HuR expression, such low levels of 5-

adenosine monophosphate-activated protein kinase (AMPK) activity (Wang et al. 2002, reviewed in López de Silanes et al. 2005). The association between

cytoplasmic HuR expression and adverse clinicopathological features of

malignant tumours such as higher grades and advanced stages has been reported

in several carcinomas such as breast, colon and ovarian carcinomas (Erkinheimo et al. 2003 Denkert et al. 2004b, Heinonen et al. 2005, Denkert et al. 2006). The

immunoexpression of HuR predicts a poor outcome, for example in breast, gastric

and ovarian carcinomas (Erkinheimo et al. 2003, Heinonen et al. 2005, Mrena et al. 2005). Moreover, the up-regulation of HuR has been found to accelerate

tumour growth in colon carcinoma animal models (López de Silanes et al. 2005).

Toll-like receptor 9

The Toll-like receptors (TLRs) are a family of recognition proteins that detect

both microbial and host-derived molecular patterns and introduce them to the

immune system (reviewed in Akira & Hemmi 2003, Wagner 2004). At least 13

different mammalian TLRs, each responding to a different ligand, have been

identified (Akira & Hemmi 2003, Shi et al. 2011). Based on amino acid sequence

similarities, the TLR family members are divided into different subfamilies. The

members of the TLR9 subfamily (TLR7, TLR8 and TLR9) recognize viral and

bacterial RNA and DNA. (reviewed in Wagner 2004) In addition, host RNA,

52

DNA and RNA- or DNA-associated proteins can stimulate TLRs and result in the

development of systemic autoimmune disease (Lamphier et al. 2006, Marshak-

Rothstein & Rifkin 2007). The subcellular expression sites of the different TLRs

vary. Some of them are expressed and bind their ligands on the cell surface

(Nishiya & DeFranco 2004), while the TLR9 subfamily resides in the intracellular

vesicles (Leifer et al. 2004). Ligand binding to TLRs launches signalling

pathways, resulting in an immune reaction (reviewed in Akira & Hemmi 2003).

The TLR9 protein is most abundantly expressed in B-cells and plasmacytoid

dendritic cells of the immune system in humans (reviewed in Wagner 2004). In

addition to the immune system, the expression of TLR9 has been shown, for

example, in normal epithelial cells of the gastric mucosa (Schmausser et al. 2004),

uterus (Schaefer et al. 2004) and respiratory epithelium (Platz et al. 2004), as well

as in keratinocytes (Mempel et al. 2003). In the normal kidney, TLR9 expression

has been detected in renal tubules and interstitial tissue. The glomerular

expression of TLR9 has been reported in autoimmune renal disease, such as in

lupus nephritis. (Papadimitraki et al. 2009)

The expression of TLR9 has also been observed in several cancers such as

breast, gastric, lung, ovarian and prostate carcinomas (Droemann et al. 2005,

Schmausser et al. 2005, Merrell et al. 2006, Ilvesaro et al. 2007, Berger et al. 2010). An association between TLR9 expression and poor differentiation of the

tumour has been described in breast, ovarian and prostate carcinomas, and also in

glioma (Jukkola-Vuorinen et al. 2008, Berger et al. 2010, Väisänen et al. 2010,

Wang et al. 2010). Stimulation of TLR9 increases the invasiveness of tumours in

several malignomas such as in breast, lung and prostate carcinomas and in glioma

(Merrell et al. 2006, Ilvesaro et al. 2007, Ren et al. 2007, Wang et al. 2010). In

breast carcinoma and glioma, TLR9 expression is correlated with an unfavourable

clinical outcome (Jukkola-Vuorinen et al. 2008, Wang et al. 2010). The molecular

mechanism via which TLR9 expression influences tumour behaviour is not fully

understood, but there is evidence linking TLRs to cell migration, invasion and

metastasis (Anderson et al. 1985, Kelleher et al. 2006, Tomchuck et al. 2008). It

has been suggested that TLR9 expression might be a part of the epithelial-to-

mesenchymal transition (EMT) and that TLR9 is involved in metastasis rather

than participating in oncogenesis itself (Jukkola-Vuorinen et al. 2008).

53

2.6.4 Neuroendocrine activity in cancer

In addition to specific neuroendocrine cells and neuroendocrine tumours,

neuroendocrine differentiation, characterized by the expression of scattered

clusters of differentiated neuroendocrine cells, the presence of secretory granules

and the production of peptide hormones or the presence of immunoreactivity for

certain neuroendocrine markers, can be detected in multiple organs and

malignomas throughout the body (reviewed in Edgren et al. 1996, Abrahamsson

1999, Erickson & Lloyd 2004). Rare neuroendocrine tumours of the kidney

include carcinoid, large cell neuroendocrine carcinoma, primitive

neuroectodermal tumour, neuroblastoma, phaeochromocytoma, small cell

carcinoma and mucinous tubular and spindle cell carcinoma of the kidney (Eble et al. 2004, Kuroda et al. 2006, Lane et al. 2007). Neuroendocrine activity is also

expressed in normal renal cells (Edgren et al. 1996).

Neuroendocrine differentiation has been observed in several types of cancer

such as breast, colorectal, lung and prostate carcinomas, as well as in RCC

(Abrahamsson et al. 1987, Gazdar et al. 1988, Nesland et al. 1988, Westlin et al. 1995). In prostate carcinoma, neuroendocrine differentiation is a common feature

that occurs in 30–100% of tumours, and it is possibly a part of the oncogenic

process (Abrahamsson et al. 1987, Abrahamsson 1999, Bostwick et al. 2002). As

the clinical behaviour of specific neuroendocrine tumours varies from aggressive

to indolent (Viallet & Ihde 1991, Dahabreh et al. 2009), the evidence of the

prognostic influence of neuroendocrine differentiation in cancer is contradictory,

too. However, it has been found that neuroendocrine products may stimulate

tumour growth and cell proliferation in an autocrine-paracrine fashion

(Abrahamsson 1999).

Serotonin

Serotonin (5-hydroxytryptamine, 5HT) is a monoamine neurotransmitter derived

from the amino acid tryptophan. It mediates diverse physiological actions in the

human body by binding to multiple receptor subtypes. (Zifa & Fillion 1992,

reviewed in Abrahamsson 1999) Serotonin is implicated in psychiatric and

neurological disorders, but it has also growth factor activity and is involved in

differentiation, cell proliferation and gene expression. In tumorigenesis, serotonin

has been connected to oncogenes, malignant transformation and tumour growth.

(Julius et al. 1989; reviewed in Vicaut et al. 2000, Siddiqui et al. 2005) The

54

tumour growth promoting effect of serotonin has been established in several types

of cancer such as lung, pancreatic and prostate carcinoma. On the other hand,

serotonin may inhibit tumour growth through its vasoconstrictive effect on

tumour vasculature. (reviewed in Vicaut et al. 2000) So far, there is no clear

evidence showing that serotonin expression is a prognostic factor in cancer. For

example, in prostate carcinoma, both increased and decreased immunoexpression

have been associated with advanced and aggressive disease (Bostwick et al. 2002,

Dizeyi et al. 2004).

CD56

Cell adhesion molecules play an import role in tumorigenesis and the

development of metastatic disease (see section 2.6.2) (reviewed in Freemont &

Hoyland 1996, Meyer & Hart 1998). The neural cell adhesion molecule (NCAM,

CD56) is a member of the immunoglobulin superfamily of cell adhesion

molecules (Jin et al. 1991). It mediates both cell-cell and cell-substrate

interactions via homophilic and heterophilic mechanisms (reviewed in Abbate et al. 1999, Daniel et al. 2001) and is a rather sensitive marker of neuroendocrine

differentiation. It is present in the central nervous system and adrenal gland and in

most neuroendocrine cells and tumours, whereas most non-endocrine cells and

tumours do not express CD56. (Jin et al. 1991, Lahr et al. 1993) The expression

of CD56 has been detected in neuroblastomas, brain tumours, embryonal

rhabdomyosarcomas and various other sarcomas, neuroendocrine tumours and

subsets of carcinomas such as lung and ovarian carcinomas, and in RCC (Garin-

Chesa et al. 1991, Terpe et al. 1993). The polysialated isoforms of CD56 are

involved in tumour invasion and metastasis (Daniel et al. 2000, Daniel et al. 2001).

Neurone-specific enolase

Enolases are cytoplasmic glycolytic enzymes consisting of three immunologically

distinct subunits α, β and γ, where subunit γ is called neurone-specific enolase

(NSE, gamma-enolase) (Fletcher et al. 1976, Schmechel et al. 1978, Ariyoshi et al. 1983). Neurone-specific enolase is a broad-spectrum, non-specific

neuroendocrine marker expressed in addition to neurons and neuroendocrine cells

in various non-endocrine cells such as myoepithelial cells and lymphocytes

(reviewed in Erickson & Lloyd 2004). In the normal kidney, NSE is expressed in

55

macula densa and in epithelial cells of loops of Henle and collecting ducts, but

not in proximal tubules. Both the up-regulation of NSE immunoexpression and

elevated serum NSE levels have been shown in RCC. (Haimoto et al. 1986) The

immunoexpression of NSE has also been observed in other cancers of non-

neuroendocrine origin, such as in breast, colorectal, lung and oesophageal

carcinomas (Ariyoshi et al. 1983, Vinores et al. 1984). The activity of glycolytic

enzymes such as NSE presumably reflects enhanced anaerobic glycolysis of the

malignant tumour. Serum NSE is regarded as a useful tumour marker for

indicating the extent of disease (Ariyoshi et al. 1983, Haimoto et al. 1986). In

RCC, increased serum levels of NSE are associated with advanced disease and

decreased survival (Takashi et al. 1989, Rasmuson et al. 1993). In contrast, in

breast carcinoma, the immunoexpression of NSE is related to improved survival

(Makretsov et al. 2003).

Chromogranin A

Chromogranins are a class of secretory proteins that reside in the secretory

granules of endocrine, neuroendocrine and neuronal cells (reviewed in Montero-

Hadjadje et al. 2008). The best characterized member of the granin family,

chromogranin A, was originally identified as a major soluble protein in adrenal

medullary chromaffin granules (Blaschko et al. 1967). Chromogranin A is

important in secretory granule formation and is thus needed in the packaging,

processing and release of a variety of hormones, neurotransmitters and peptides in

response to physiological and pharmacological stimuli. It is a precursor to several

bioactive peptides that have diverse autocrine, paracrine and endocrine activities.

Chromogranin A and its peptides play multiple roles in the endocrine,

cardiovascular and nervous systems and they are implicated in cardiovascular and

neurodegenerative diseases. In addition, chromogranin A has apoptosis

modulating activity. (reviewed in Taupenot et al. 2003, Montero-Hadjadje et al. 2008)

Chromogranin A immunoexpression has been demonstrated in several non-

endocrine tumours, including breast, colorectal, gastric, lung and prostate

carcinomas (Abrahamsson et al. 1989, Pagani et al. 1990, Skov et al. 1991, Park et al. 1992). Increased serum levels of chromogranin A have been reported in

patients with colorectal, prostate and renal cell carcinoma (Kadmon et al. 1991,

Syversen et al. 1995, Rasmuson et al. 1999). Chromogranin A is regarded as a

potential tumour marker for several types of neoplasms (reviewed in Taupenot et

56

al. 2003). Although the prognostic significance of chromogranin A expression

remains unclear in cancer, a tendency towards shorter survival has been reported

in colorectal carcinoma (Syversen et al. 1995, reviewed in Taupenot et al. 2003).

Synaptophysin

Neurons, as well as neuroendocrine and endocrine cells, contain small presynaptic

secretory vesicles that store and release low molecular weight components such

as classic neurotransmitters. The expression of a protein called synaptophysin at

the cytoplasmic site of the presynaptic vesicle membrane is characteristic of these

small neurosecretory vesicles. (reviewed in Wiedenmann & Huttner 1989)

Synaptophysin is a calcium-binding transmembrane glycoprotein involved in

forming the transmembrane channels needed in the accumulation and release of

vesicular contents (reviewed in Wiedenmann & Huttner 1989, Erickson & Lloyd

2004). It is a marker of nerve terminals and a broad-spectrum marker of

neuroendocrine differentiation (reviewed in Erickson & Lloyd 2004). In addition

to neurons and many endocrine and neuroendocrine cells, the immunoexpression

of synaptophysin has been reported in various cancers, for example in breast,

endometrial and lung carcinomas, with no significant association with patient

survival (Kayser et al. 1988, Makretsov et al. 2003, Taraif et al. 2009; reviewed

in Wiedenmann & Huttner 1989, Erickson & Lloyd 2004).

57

3 Aims of the study

The purpose of this thesis was to screen renal cell carcinoma tumours with

different immunohistochemical biomarkers to find novel prognostic tissue

markers for RCC. Specifically, the aims of the study were:

1. to investigate the expression of markers of the oxidative system, including

Keap1, Nfr2 and Trx, in RCC tumours and their correlation with

clinicopathological parameters such as TNM classification and nuclear grade

and survival in RCC;

2. to evaluate the expression of myosin VI and its prognostic significance in

RCC and to discover how myosin VI expression relates to the expression of

E-cadherin and beta-catenin and clinicopathological features in RCC;

3. to study the expression and prognostic role of the mRNA stability factor HuR

in RCC and to investigate the presence of HuR-regulated COX-2 expression

in RCC;

4. to explore the expression of TLR9 in RCC and to determine whether or not it

is correlated with patient RCC-specific survival as well as with TNM stage

and nuclear grade of RCC tumours;

5. to discover the expression of neuroendocrine markers in RCC and their

influence on RCC-specific patient survival.

58

59

4 Materials and methods

4.1 Study cohort

The retrospective study population was based on a cohort of 189 patients who

underwent renal surgery for renal tumours at the Department of Surgery, Oulu

University Hospital, Finland, between 1990 and 1999, and who were identified

from the registry of the Department of Pathology, Oulu University Hospital. On

the basis of histologically verified RCC and the availability of representative

histological materials and follow-up details, 152 patients were included in the

study.

4.2 Clinical data and follow-up

All relevant clinical data and follow-up details were collected from patient

records and evaluated along with information from the Finnish Cancer Registry

and the Population Register Centre of Finland. All information used was re-

evaluated. The presence of chronic diseases, including hypertension and other

cardiovascular diseases, the use of antihypertensive medication and smoking and

RCC-related symptoms were recorded. Because of the retrospective nature of the

study, some information, for example alcohol use, which was not documented,

was not available. Performance status could not be retrospectively assessed based

on the documentation. The preoperative laboratory findings and radiological

evaluation, as well as information regarding surgical treatment and peri- and

postoperative complications, were carefully recorded. Adjuvant oncologic

treatment was also recorded. Follow-up details were meticulously collected from

each patient as required from separate hospitals and health care centres all over

Finland.

The exact stage was recorded according to the TNM classification of RCCs

2002 (Sobin & Wittekind 2002) (Figure 1), which was the version in clinical use

in 2006 when the study began. In order to simplify the statistical analyses the

TNM stages were further grouped into stages I-IV (Figure 1). The lymph node

status was determined via pathological evaluations of resected lymph nodes and

was classified N0 when no lymph node metastases were histologically verified. In

a case of histologically verified lymph node metastasis, the tumour was staged N1

when there was only one positive regional lymph node or N2 when there were

60

more than one regional lymph node metastases, even if the number of lymph

nodes analysed was not documented or if fewer than eight lymph nodes were

analysed which is the minimum number required for TNM classification.

4.3 Tumour samples

All of the tumour samples were specimens from nephrectomy or partial resection.

The tumour samples were routinely fixed in 10% buffered formalin and

embedded in paraffin. The histological diagnosis was confirmed by reviewing the

original haematoxylin and eosin (H & E)-stained sections by two

histopathologists (Pasi Hirvikoski, M.D., Ph.D. and Saila Kauppila M.D., Ph.D.).

The tumours were re-classified and graded according to the current WHO

classification (Eble et al. 2004). Microscopic venous invasion (MVI), tumour

necrosis, sarcomatoid features and cystic structure were also assessed.

4.4 Multi-tissue blocks

The most representative tumour region of each RCC tumour was defined and

marked in the H & E-stained sections. The regions marked were punched from

the original blocks and re-cast in multitissue blocks. Sections of 3 μm thick were

cut from these paraffin-embedded blocks and mounted onto pre-coated slides

(tissue microarray technique, TMA) (Merseburger et al. 2006).

4.5 Immunohistochemistry

Tissue sections were deparaffinized in xylene, rehydrated in a descending ethanol

series and washed in phosphate buffered saline (PBS). In order to enhance

immunoreactivity the tissue sections were boiled in citrate buffer (pH 6.0) or in

Tris-EDTA buffer (pH 9.0), after which they were cooled for 20 min in a buffer.

Endogenous peroxidase activity was blocked by incubation in methanol-

containing hydrogen peroxidase. The sections were incubated with specific

antibodies and the antibody reaction was visualized using commercial kits. The

details of the antibodies, specific pretreatments and visualization methods used

are listed in Table 1. All steps in the procedure were carried out at room

temperature. The sections were washed with PBS in between each step in the

staining procedure.

61

Table 1. Details of the immunohistochemical procedure.

Marker Antibody;

Manufacturer

Dilution Pretreatment Visualization;

Chromogen

Nfr2 Rabbit polyclonal antibody, Nfr2 C-

20: sc-722;

Santa Cruz Biotechology, Inc.

1/100 Citrate1 Histostain Plus,

Invitrogen Life

Technologies;

AEC4

Keap1 Goat polyclonal IgG antibody, Keap-

T E-20 SC-15246;

Santa Cruz Biotechology, Inc.

1/100 Citrate1 Histostain Plus,

Invitrogen Life

Technologies;

AEC4

Trx Goat anti-human IgG antibody,

populi No 705;

American Diagnostica, Inc.

1/200 Citrate1 Histostain Plus,

Invitrogen Life

Technologies¸;

AEC4

Myosin VI Monoclonal anti-myosin VI;

Sigma

1/250 Citrate1, TBS3 EnVision,

DakoCytomation;

DAB5

E-cadherin Mouse monoclonal anti-E-cadherin;

Zymed Laboratories

1/300 Citrate1, TBS3 EnVision,

DakoCytomation;

DAB5

β-catenin Monoclonal anti-beta-catenin;

BD Biosciences

1/200 Citrate1, TBS3 EnVision,

DakoCytomation;

DAB5

HuR Mouse monoclonal antibody HuR,

19F12, sc-56709;

Santa Cruz Biotechnology, Inc.

1/1500 Tris-EDTA2 EnVision,

DakoCytomation;

DAB5

COX-2 Mouse monoclonal antibody COX-2;

Cayman Chemical

1/200 Citrate1 EnVision,

DakoCytomation;

DAB5

TLR9 Mouse monoclonal anti-human

TLR9/CD289, Img-305A, clone

26C593.2;

Imgenex

1/200 Citrate1 EnVision,

DakoCytomation;

DAB5

Serotonin Mouse monoclonal anti-human

serotonin;

DakoCytomation

1/200 Citrate1 UltraVision, Thermo

Fisher Scientific;

DAB5

CD56 Mouse lyophilized monoclonal

antibody for CD56;

Novocastra Laboratories Ltd.

1/200 Citrate1 UltraVision, Thermo

Fisher Scientific;

DAB5

62

Marker Antibody;

Manufacturer

Dilution Pretreatment Visualization;

Chromogen

NSE Mouse monoclonal anti-NSE;

Zymed Laboratories

1/1000 None UltraVision, Thermo

Fisher Scientific;

DAB5

Chromogranin A Rabbit polyclonal anti-chromogranin

A;

Zymed Laboratories

1/500 Tris-EDTA2 EnVision,

DakoCytomation;

DAB5

Synaptophysin Mouse monoclonal anti-

synaptophysin;

DakoCytomation

1/50 Tris-EDTA2 EnVision,

DakoCytomation;

DAB5 1Boiled in citrate buffer (pH 6.0) in a microwave oven for 2 min at 800 W + 10 min at 300 W 2Boiled in Tris-EDTA buffer (pH 9.0) in a microwave oven for 2 min at 800 W + 15 min at 300 W 3Washed with Tris-buffered saline (TBS) (pH 7.5) 43-amino-9-ethylcarbazole 53–3'-diaminobenzidine

4.6 Evaluation of immunohistochemial staining

In addition to internal controls, additional positive controls were used when

appropriate. Sections that were stained without primary antibodies served as

negative controls. Immunoreactivity was simultaneously assessed by two to four

interpreters who had no preliminary knowledge of any clinical data and a

consensus was reached.

4.6.1 Nfr2 (I)

The immunostaining of Nfr2 was observed in both the nucleus and the cytoplasm.

Immunoreactivity was classified into four categories representing subgroups of

negative, weakly positive, moderately positive and strongly positive staining

reactions. The limits for these categories were 0%, < 20% positive, 20–40%

positive and > 40% positive for nuclear Nfr2 and 0%, < 20% positive, 20–50%

positive and > 50% positive for cytoplasmic Nfr2.

4.6.2 Keap1 (I)

Immunoreactivity for Keap1 was detected in the cytoplasm. Immunostaining for

Keap1 was primarily scored into four subgroups based on the intensity of staining:

negative (0), weakly positive (< 10%), moderately positive (10–40%) and

63

strongly positive (> 40%). In a further statistical analysis the positive subgroups

were combined and compared to the negative subgroup.

4.6.3 Thioredoxin (I)

Immunostaining for nuclear and cytoplasmic Trx was classified into four groups

according to the staining reaction: negative (0), weakly positive (< 25%),

moderately positive (25–75%) and strongly positive (> 75%). Statistical analyses

were performed between positive and negative cases for cytoplasmic Trx.

4.6.4 Myosin VI (II)

Immunoreactivity for myosin VI was observed in the nucleus and cytoplasm.

Nuclear myosin VI immunoreactivity was classified as positive when at least one

stained nucleus was detected and negative when no stained nuclei were detected.

Immunostaining for cytoplasmic myosin VI was initially scored as negative,

weakly positive, moderately positive and strongly positive to represent the

intensity of staining. In a further statistical analysis the negative and weakly

positive subgroups were combined and re-classified as negative and compared to

the subgroups of moderately and strongly positive, which were combined and re-

classified as positive.

4.6.5 E-cadherin (II)

Immunoreactivity for E-cadherin was detected in the nucleus and membranes.

Immunostaining for nuclear E-cadherin was classified as positive when staining

was detected and negative if no staining was detected. Immunostaining for

membranous E-cadherin was primarily categorized as negative, weakly positive,

moderately positive and strongly positive based on the intensity of the staining

reaction. In the final analyses, immunostaining for membranous E-cadherin was

re-classified as negative (negative and weakly positive combined) or positive

(moderate and strongly positive combined).

64

4.6.6 Beta-catenin (II)

Immunoreactivity for beta-catenin was observed in the nucleus and cytoplasm and

classified as positive when staining was detected or negative when no staining

was detected.

4.6.7 HuR (III)

Immunoreactivity for HuR was detected in the nucleus and cytoplasm. Nuclear

immunoreactivity was scored as either negative or positive. Cytoplasmic HuR

immunoreactivity was initially scored depending on the cytoplasmic intensity of

staining: negative (0), weakly positive (1) or strongly positive (2). For further

statistical analyses the negative samples (score 0) were compared to the positive

samples (scores 1 and 2).

4.6.8 Cyclooxygenase-2 (III)

Cytoplasmic COX-2 immunoreactivity was assessed by grading the intensity of

staining from 0 to 3 (absent, weakly diffuse, moderately granular, strong granular

or diffuse, respectively). In the statistical analysis, the COX-2 grades were

divided into two subgroups, i.e. negative (scores 0–1) and positive (scores 2–3).

4.6.9 Toll-like receptor 9 (IV)

Immunoreactivity of TLR9 was detected in both the nucleus and cytoplasm.

Nuclear immunostaining was assessed as negative or positive. Cytoplasmic TLR9

immunoreactivity was initially scored according to four cytoplasmic staining

intensities: negative (0), weakly positive (1), moderately positive (2) or strongly

positive (3). For further statistical analyses, the negative samples (score 0) were

compared with the positive ones (scores 1 to 3).

4.6.10 Neuroendocrine markers (V)

Cytoplasmic immunostaining for serotonin, CD56, NSE, chromogranin A and

synaptophysin was classified as positive when any staining was detected or

negative when no staining was detected.

65

4.7 Cell culture and RNA interference (III)

Human RCC cells (769-P, ATCC CRL-1933) were cultured in Roswell Park

Memorial Institute medium (RPMI-1640) supplemented with 10% foetal calf

serum (FCS) (Sigma-Aldrich), 2 mmol/l L-glutamine and antibiotics (Sigma-

Aldrich) and maintained at 37 °C and 5% carbon dioxide (CO2) in air. The small

interfering RNA (siRNA) duplexes were synthesized by Dharmacon, Inc. and the

sequences for HuR were: sense 5′-AAC AUG ACC CAG GAU GAG UUA dTdT-

3′ and antisense 5′-UAA CUC AUC CUG GGU CAU GUU dTdT-3′. The ON-

TARGETplus Non-targeting Pool (D-001810-10, Dharmacon) was used as a

control for the siRNA experiments. The day before transfection, 769-P cells were

trypsinized and diluted at 1:20 with optiMEM 1-medium (Life Technologies, Inc.)

supplemented with 10% FCS without antibiotics and transferred to 24-well plates

at 1 ml per well with a final split ratio of 1:4. Transient transfection of the siRNAs

was carried out using Lipofectamine reagent (Invitrogen Life Technologies),

following the manufacturer’s instructions. The final siRNA concentrations were

80 nmol/l.

4.8 Immunoblotting (III)

Proteins from the cells were isolated 78 h after transfection using a

radioimmunoprecipitation assay (RIPA) buffer containing 25 mM Tris-

hydrochloride (Tris-HCl) (pH=7.6), 150 mM sodium chloride (NaCl), 1% nonyl

phenoxypolyethoxylethanol-40 (NP-40), 1% sodium deoxycholate and 0.1%

sodium dodecyl sulphate (SDS). After boiling the samples in a reducing SDS

sample buffer for 5 min, 20 μg of protein was loaded per lane and the samples

were separated by 10% SDS-PAGE. Proteins from the SDS gel were transferred

to a polyvinylidene fluoride (PVDF) membrane for HuR (sc-56709, Santa Cruz

Biotechnology), COX-2 (sc-1745, Santa Cruz Biotechnology) and glyceraldehyde

3-phosphate dehydrogenase (GAPDH) (SM7116P, Acris Antibodies GmbH)

detection. Binding of target proteins on the membranes was revealed with goat

anti-mouse or donkey anti-goat horseradish peroxidase (HRP)-conjugated

secondary antibodies (Santa Cruz Biotechnology). The protein bands were

visualized by chemiluminescence.

66

4.9 Statistical analyses

The software package SPSS for Windows 15 was used for the statistical analysis.

Associations between factors, including clinicopathological variables and the

immunostaining pattern, were assessed using the chi-square test (χ2 test) or

Fisher’s exact test in the case of low expected frequencies. Survival analyses were

performed for the whole study population and also separately for each

histological subtype. Survival rates were calculated using the Kaplan-Meier

method and statistical significance between the groups was analysed using the

log-rank test. The hazard ratio (HR) was assessed using a Cox regression model

and univariate analysis. The RCC-specific survival was calculated from the date

of diagnosis to death due to RCC or to the last day of follow-up. Deaths due to

intercurrent causes were not included in the analysis. Multivariate survival

analysis was performed using the Cox proportional hazards model with the

covariates age, stage, grade and immunoreactivity for each biomarker tested. All

of the p-values were two-sided.

4.10 Ethics

The research plan was approved by the Ethics committee of the Northern

Ostrobothnia Hospital District (EETTMK:16/2005). Approval for the study was

sought from the Finnish Ministry of Social Affairs and Health (STM/1117/2005)

in order to obtain data from the Finnish Cancer Registry. The use of pathological

archive material was licenced by the National Authority for Medicolegal Affairs

(at the present time the National Supervisory Authority for Welfare and Health)

(Permission No. 1441/32/300/05).

67

5 Results

5.1 Characteristics, treatment and follow-up of the patients

The mean and median age of the patients were 62 (standard deviation (SD) 11)

and 63 (range 29–86) years, respectively. There was a slight predominance of

females in the study cohort with 77 (51%) females and 75 (49%) males. Elevated

blood pressure, the use of antihypertensive medication or a documented

hypertension diagnosis was observed in 73 (48%) of the patients. There were 32

(21%) smokers in the study cohort but the smoking anamnesis was missing in 18

(12%) cases.

Approximately one half (77 tumours, 51%) of the tumours had been found

incidentally. A preoperative abdominal US had been performed for each patient

and, in addition, an abdominal CT was performed for 125 patients (82%).

Preoperative chest radiography (X-ray and/or CT) was performed for 135 patients

(89%). In the case of suspected metastases or vena caval involvement, additional

preoperative studies such as bone scintigraphy (14 patients, 9%), skeletal

radiography (17 patients, 11%), MRI (11 patients, 7%) or cavography (3 patients,

2%) were performed. Preoperative embolization was successfully performed in 37

(24%) of the tumours. Out of the patients, 6 (4%) had histologically verified

lymph node metastases and 18 (12%) had distant metastases at the time of

primary surgical treatment. The most common organ sites for metastases were

lung, bone and liver, representing 7 (39%), 6 (33%) and 3 (17%) of metastases at

the time of diagnosis, respectively. In addition, there were solitary metastases in

the adrenal gland, brain, uterus, oment and subcutis. Three of the patients (2%)

had metastases in more than one organ site. In the study cohort, 70 (46%), 12

(8%), 51 (34%) and 19 (12%) patients had stages I to IV, respectively. The

majority of patients underwent radical nephrectomy (145 patients, 95%) and 7

(5%) partial nephrectomy. The tumour characteristics are presented in Table 2.

The median and mean times of follow-up were 90 (range 0–209, SD 63.6)

months. The follow-up was complete in all cases. During the follow-up period, 44

(29%) patients died of RCC, 40 (26%) died of other causes and 68 (45%) were

still alive. Thirty-two patients suffered from RCC-metastases during the follow-

up period. Metastases were diagnosed in 8 patients within one year after the

primary operation and in 6, 8, 5 and 5 patients after 1–3 years, 3–5 years, 5–10

years and after more than 10 years of follow-up, respectively.

68

Table 2. Tumour characteristics.

Characteristic Occurrence

Side

Right 76/152 (50%)

Left 75/152 (49%)

Bilateral 1/152 (1%)

Tumour class

T1 75/152 (49%)

T2 12/152 (8%)

T3 59/152 (39%)

T4 6/152 (4%)

Nodal status

N0 146/152 (96%)

N1 4/152 (3%)

N2 2/152 (1%)

Distant metastases

M0 134/152 (88%)

M1 18/152 (12%)

Stage

I 70/152 (46%)

II 12/152 (8%)

III 51/152 (34%)

IV 19/152 (12%)

Histology

Clear cell 134/152 (88%)

Papillary 11/152 (7%)

Chromophobic 5/152 (3%)

Undifferentiated 2/152 (1%)

Nuclear grade

I 5/150 (3%)

II 83/150 (55%)

III 40/150 (27%)

IV 22/150 (15%)

5.2 Survival analysis of the clinicopathological characteristics

The 5- and 10-year RCC-specific survival and HR for deaths due to RCC,

according to the major clinicopathological parameters are shown in Table 3.

69

The RCC-specific survival rate did not differ between males and females, age

groups or symptomatic and asymptomatic tumours in univariate or multivariate

analysis. Of the various preoperative symptoms reported, hypogastric pain (14

patients, 9%) had a prognostic value in RCC-specific survival, with p=0.02.

Preoperative anaemia, hypersedimentation and increased CRP were markers of a

poorer RCC-specific prognosis, with p=0.002, p=0.004 and p<0.001, respectively,

in univariate but not in multivariate survival analysis. None of the patients with

normal CRP levels died from RCC during the follow-up period. Other laboratory

findings such as increased liver enzyme levels, polycythaemia and

hypercalcaemia had no prognostic significance in this material (data not shown).

Time to metastasis was a statistically significant prognostic factor in RCC-

specific survival in both the univariate (p<0.001) and multivariate analysis

(p=0.004) adjusted for age, stage and grade.

The TNM classification, stage, tumour size and several histological

parameters including nuclear grade, sarcomatoid features, microscopic venous

invasion (MVI) and necrosis were statistically significant prognostic factors in the

univariate analysis. Although papillary and chromophobic RCCs had a slightly

longer RCC-specific mean survival (155 (95% CI 124–185) and 169 (95% CI

105–234) months, respectively) than clear cell RCCs (150 (95% CI 135–165)

months), the difference was not statistically significant (p=0.7). Cystic structure

was not a prognostic factor in our material. In the multivariate analysis adjusted

for age, stage and grade, the only independent prognostic factors of the above-

mentioned clinicopathological parameters were TNM staging, stage and tumour

size (see Table 3).

70

Table 3. The 5- and 10-year RCC-specific survival (Kaplan-Meier method) and HR for

deaths due to RCC (Cox univariate analysis) according to the clinicopathological

parameters measured.

Parameter 5-year survival 10-year survival HR 95% CI p-value

All 77% 72%

Gender

Male 73% 71% 1.30 0.72–2.35 0.4

Female 81% 74% 1 (ref.)

Age

Under 65 81% 75% 1 (ref.)

65 or over 72% 69% 1.47 0.81–2.66 0.2

Under 46 80% 80% 1 (ref.)

46 or over 77% 72% 1.23 0.38–4.00 0.7

Presentation

Asymptomous 78% 75% 1 (ref.)

Symptomous 76% 69% 1.41 0.78–2.56 0.3

T class

T1 92% 88%

T1a 98% 93% 1 (ref.)

T1b 86% 83% 3.16 0.82–12.24 0.1

T2 75% 67% 6.96 1.66–29.17 0.008

T3 64% 59%

T3a 68% 65% 6.03 1.72–21.20 0.005

T3b 67% 61% 7.27 1.88–28.16 0.004

T3c 25% 0% 32.95 7.27–149.27 <0.001

T4 17% 17% 26.74 6.32–113.06 <0.001

Nodal status

N0 79% 74% 1 (ref.)

N1 50% 50% 3.70 0.89–15.41 0.07

N2 0% 0% 42.64 8.45–215.08 <0.001

Distant metastasis

M0 86% 81% 1 (ref.)

M1 11% 11% 13.75 6.91–27.37 <0.001

Stage

I 96% 91% 1 (ref.)

II 75% 67% 5.12 1.62–16.17 0.005

III 75% 69% 3.80 1.56–9.24 0.003

IV 16% 16% 24.09 9.61–60.41 <0.001

71

Parameter 5-year survival 10-year survival HR 95% CI p-value

Tumour size (cm)

Under 5 92% 86% 1 (ref.)

From 5 to 9 74% 70% 2.65 1.15–6.14 0.02

10 or over 50% 45% 6.70 2.62–17.13 <0.001

Nuclear grade

I 100% 100% 1 (ref.)

II 88% 82% 1 (ref.)

III 68% 68% 1.76 0.86–3.63 0.1

IV 50% 41% 3.92 1.90–8.09 <0.001

MVI

No 80% 75% 1 (ref.)

Yes 47% 47% 3.89 1.86–8.16 <0.001

Sarcomatoid

No 79% 76% 1 (ref.)

Yes 50% 33% 3.79 1.75–8.23 0.001

Necrosis

No 88% 84% 1 (ref.)

Yes 60% 55% 3.62 1.96–6.71 <0.001

Histology

Clear cell 76% 71% 1 (ref.)

Papillary 91% 91% 0.60 0.14–2.48 0.5

Chromophobic 80% 80% 0.58 0.08–4.23 0.6

Undifferentiated 50% 50% 2.31 0.32–16.84 0.4

Cystic structure

No 72% 69% 1.73 0.59–5.01 0.3

Yes 89% 83% 1 (ref.)

MVI microscopic venous invasion

1 (ref.) hazard ratio 1.0, the reference group

5.3 Staining results and their association with the clinico-pathological parameters measured and disease-specific

survival

The proportion of positive immunostaining and associations between tumour

markers and the clinicopathological characteristics measured, including stage,

TNM classification, histology and nuclear grade, at the time of diagnosis are

presented in Table 4. The 5- and 10-year RCC-specific survival rates (Kaplan-

72

Meier analysis) and HR for deaths due to RCC (Cox univariate analysis),

according to the expression of tumour markers, are presented in Table 5.

Table 4. Proportion of positively stained specimens and the relationship between the

tumour markers and the clinicopathological characteristics.

Marker Positive/all Stage T N M Histology Grade

Nuclear Nfr2 113/139 (81%) 0.007 0.005 NS NS NS 0.02

Cytoplasmic Nfr2 135/139 (97%) NS NS NS NS NS NS

Keap1 79/144 (55%) <0.001 0.002 NS 0.002 NS NS

Nuclear Trx 111/143 (78%) NS NS NS NS NS 0.02

Cytoplasmic Trx 123/143 (86%) NS NS NS NS NS NS

Nuclear myosin VI 51/145 (35%) NS NS NS NS NS NS

Cytoplasmic myosin VI 104/145 (72%) NS NS NS NS NS 0.04

Nuclear E-cadherin 59/148 (40%) NS NS NS NS NS NS

Membranous E-cadherin 14/148 (9%) NS NS NS NS <0.001 NS

Nuclear β-catenin 65/147 (44%) NS NS NS NS NS 0.003

Cytoplasmic β-catenin 13/147 (9%) NS NS NS NS NS NS

Nuclear HuR 127/145 (88%) NS NS NS NS NS NS

Cytoplasmic HuR 37/145 (25%) 0.004 0.04 NS 0.009 0.01 NS

Cytoplasmic COX-2 22/147 (15%) 0.04 0.03 0.04 0.02 0.007 0.04

Cytoplasmic TLR9 112/138 (81%) NS NS NS NS NS NS

Serotonin 12/145 (8%) NS NS NS NS NS NS

CD56 26/144 (18%) NS NS NS NS NS NS

NSE 69/145 (48%) NS NS NS NS 0.01 NS

Synaptophysin 2/148 (1%) NS NS NS NS NS NS

Chromogranin A 0/148 (0%)

NS no significant correlation; numeric value is p-value for correlation

Table 5. The 5- and 10-year RCC-specific survival rates (Kaplan-Meier method) and HR

for deaths due to RCC (Cox univariate analysis) according to tumour marker

expression.

Marker 5-year survival 10-year survival HR 95% CI p-value

Nuclear Nfr2

Negative 73% 65% 1.09 0.36–3.27 0.9

Weakly positive 85% 80% 0.69 0.25–1.92 0.5

Moderately positive 68% 66% 1.17 0.42–3.29 0.8

Strongly positive 64% 64% 1 (ref.)

Cytoplasmic Nfr2

Negative 100% 75% 1 (ref.)

Weakly positive 80% 78% 1.08 0.14–8.35 0.9

Moderately positive 77% 73% 1.47 0.19–11.08 0.7

73

Marker 5-year survival 10-year survival HR 95% CI p-value

Strongly positive 67% 60% 1.75 0.23–13.44 0.6

Cytoplasmic Keap1

Negative 88% 80% 1 (ref.)

Positive 66% 63% 2.09 1.11–3.94 0.02

Nuclear Trx

Negative 78% 69% 1.49 0.62–3.60 0.4

Weakly positive 76% 69% 1.53 0.62–3.78 0.4

Moderately positive 72% 70% 1.33 0.57–3.10 0.5

Strongly positive 78% 75% 1 (ref.)

Cytoplasmic Trx

Negative 70% 55% 1.87 0.90–3.88 0.1

Positive 76% 73% 1 (ref.)

Nuclear myosin VI

Negative 78% 72% 1 (ref.)

Positive 75% 71% 1.04 0.56–1.95 0.9

Cytoplasmic myosin VI

Negative 83% 78% 1 (ref.)

Positive 74% 69% 1.40 0.69–2.85 0.3

Nuclear E-cadherin

Negative 74% 70% 1.31 0.7–2.43 0.4

Positive 80% 75% 1 (ref.)

Membranous E-

cadherin

Negative 78% 73% 1 (ref.)

Positive 57% 57% 1.59 0.67–3.77 0.3

Nuclear β-catenin

Negative 72% 68% 1.34 0.73–2.48 0.3

Positive 82% 75% 1 (ref.)

Cytoplasmic β-catenin

Negative 78% 72% 1 (ref.)

Positive 62% 62% 1.41 0.56–3.59 0.5

Nuclear HuR

Negative 94% 83% 1 (ref.)

Positive 76% 72% 1.83 0.65–5.16 0.3

Cytoplasmic HuR

Negative 83% 78% 1 (ref.)

Positive 62% 60% 2.18 1.16–4.09 0.02

74

Marker 5-year survival 10-year survival HR 95% CI p-value

COX-2

Negative 81% 76% 1 (ref.)

Positive 55% 50% 2.29 1.15–4.54 0.02

Cytoplasmic TLR9

Negative 62% 58% 2.40 1.24–4.63 0.009

Positive 80% 75% 1 (ref.)

Serotonin

Negative 75% 70% 2.08 0.50–8.60 0.3

Positive 83% 83% 1 (ref.)

CD56

Negative 80% 75% 1 (ref.)

Positive 65% 62% 1.37 0.67–2.79 0.4

NSE

Negative 80% 72% 1 (ref.)

Positive 73% 71% 1.09 0.60–1.99 0.8

1 (ref.) hazard ratio 1.0, the reference group

5.3.1 Markers of the oxidative system (I)

Nfr2

Immunoreactivity for Nfr2 was investigated in 139 RCC specimens. Nuclear Nfr2

immunoreactivity was observed in 113 (81%) RCC tumours, of which 61 (44%)

were weakly positive, 38 (27%) were moderately positive and 14 (10%) were

strongly positive. Cytoplasmic Nfr2 was positive in 135 (97%) cases, of which 50

(36%) were weakly positive, 52 (37%) were moderately positive and 33 (24%)

were strongly positive.

Nuclear Nfr2 immunoexpression was associated with stage (p=0.007),

tumour (T) class (p=0.005) and grade (p=0.02), and cytoplasmic Nfr2 with age

(p=0.007). Nfr2 immunoreactivity did not have any prognostic influence on RCC-

specific survival either in uni- or multivariate analysis.

Keap1

The staining pattern of Keap1 was cytoplasmic. Immunoreactivity of Keap1 was

evaluated in 144 RCC tumours, of which 79 (55%) were immunopositive. Of the

positive cases, 44 (31%), 24 (17%) and 11 (8%) tumours were weakly,

75

moderately and strongly positive, respectively. In the statistical analysis the

immunopositive tumours were grouped together and compared with the

immunonegative ones.

The immunoreactivity of Keap1 was associated with several factors

indicating the advanced pattern of RCC, including advanced stage (p<0.001),

a higher T class (p=0.002) and presence of distant metastases (p=0.002). The

higher nuclear grade RCC tumours were more frequently immunopositive for

Keap1, although the difference was not statistically significant (p=0.07).

In the univariate analysis, Keap1 immunoreactivity correlated with reduced

RCC-specific survival, with p=0.02. The mean RCC-specific survival was 167

(95% confidence interval (CI) 148–186) months for Keap1-immunonegative

tumours and 133 (95% CI 113–154) months for Keap1-immunopositive tumours.

When Keap1 immunoreactivity was subjected to multivariate analysis adjusted

for age, stage and grade, the only independent prognostic factor was stage.

Thioredoxin

There were 143 RCC samples available for Trx immunohistochemistry. Nuclear

Trx immunoreactivity was detected in 111 (78%) tumours, of which 29 (20%)

were weakly positive, 46 (32%) were moderately positive and 36 (25%) were

strongly positive. Cytoplasmic Trx immunostaining occurred in 123 (86%) cases:

14 (10%) were weakly positive, 38 (27%) were moderately positive and 71 (50%)

were strongly positive. Statistical analyses were performed between

immunopositive and immunonegative tumours for cytoplasmic Trx.

The immunoreactivity of Trx was more frequently observed in well-

differentiated tumours than in poorly differentiated ones. Nuclear and cytoplasmic

Trx immunoreactivity was associated with nuclear grade, with p=0.02 and p=0.05,

respectively.

The tumours with cytoplasmic immunoreactivity of Trx had a tendency

toward a better RCC-specific prognosis as the mean RCC-specific survival was

115 (95% CI 78–153) months for tumours with no cytoplasmic expression of Trx

and 153 (95% CI 138–169) months for tumours with cytoplasmic expression of

Trx. However, the difference was not statistically significant (p=0.09). Nuclear

Trx was not associated with RCC-specific survival. In the multivariate analysis

adjusted for age, stage and grade, Trx immunoreactivity was not found to be a

prognostic factor.

76

Associations between oxidative markers

Cytoplasmic Nfr2 was associated with nuclear Nfr2 (p<0.001) and cytoplasmic

Keap1 (p=0.009). Nuclear Nfr2 was correlated with Trx immunoreactivity in both

the nucleus and the cytoplasm (p=0.01 and p=0.03, respectively). In addition,

nuclear and cytoplasmic Trx were associated with each other (p<0.001).

5.3.2 Myosin VI, E-cadherin and beta-catenin (II)

Myosin VI

Immunoreactivity for myosin VI was assessed in 145 RCC samples. Nuclear

immunostaining was observed in 51 (35%) tumours. Cytoplasmic myosin VI was

primarily detected in 28 (19%), 50 (35%) and 54 (37%) tumours representing

subgroups of weak, moderate and strong immunopositivity, respectively, whereas

13 tumours (9%) were immunonegative for myosin VI. In a further analysis the

strongly and moderately positive tumours were combined and regarded as

positive (104 tumours, 72%) and compared to the negative ones (41 weakly

positive and negative tumours, 28%).

Nuclear myosin VI was not associated with any of the clinicopathological

parameters tested. Cytoplasmic myosin VI was associated with a lower Fuhrman

grade, with p=0.04.

The mean RCC-specific survival was 162 (95% CI 137–187) months for

cytoplasmic myosin VI-negative tumours and 146 (95% CI 128–163) months for

positive tumours (p=0.3). The survival difference was observed also in low grade

tumours (data not shown). In the multivariate analysis adjusted for age, stage and

grade, cytoplasmic myosin VI immunoexpression was found to be an independent

prognostic factor along with stage. The HR for cytoplasmic myosin VI-expressing

tumours was 2.3 (95% CI 1.1–5.0, p=0.03) when compared to the

immunonegative tumours. In the multivariate analysis for clear cell carcinomas,

cytoplasmic myosin VI expression was almost a statistically significant (p=0.05)

prognostic factor of poorer prognosis (HR 2.1, 95% CI 1.0–4.5). Nuclear myosin

VI expression was not correlated with RCC-specific survival.

77

E-cadherin

There were 148 RCC tumours specimens available to evaluate E-cadherin

immunoreactivity. Immunostaining for nuclear E-cadherin was observed in 59

(40%) tumours. Membranous E-cadherin was detected as follows: weak intensity

in 27 (18%), moderate intensity in 8 (5%) and strong intensity in 6 (4%) of the

tumours. Moderately and strongly positive tumours (14, 9%) were combined and

re-classified as positive and the negative and weakly positive tumours were

combined and re-classified as negative (134, 91%).

No relationships were found between nuclear E-cadherin expression and

clinicopathological features. Immunoreactivity for membranous E-cadherin was

associated with histology (p<0.001) and was more frequently detected in

chromophobic and undifferentiated carcinomas than in clear cell tumours. None

of the papillary carcinomas showed membranous E-cadherin immunopositivity.

Immunostaining for E-cadherin in either the nucleus or membranes was not a

prognostic factor in RCC-specific survival.

Beta-catenin

Of 147 RCC tumours, 65 (44%) were immunopositive for nuclear beta-catenin.

Cytoplasmic beta-catenin was observed in 13 out of 147 tumours (9%).

Nuclear beta-catenin immunoreactivity was associated with a lower Fuhrman

grade (p=0.003). Cytoplasmic beta-catenin was not associated with any of the

clinicopathological parameters tested.

The uni- and multivariate analyses of the whole study population did not find

beta-catenin immunoexpression to be a prognostic factor. However, when the

survival analysis was performed separately for the subgroup of clear cell

carcinomas, cytoplasmic beta-catenin expression was almost a statistically

significant independent marker of decreased survival in the multivariate analysis

adjusted for age, stage and grade, with an HR of 3.0 for positive tumours

compared to negative ones (95% CI 1.0–9.0, p=0.05).

Associations between myosin VI, E-cadherin and beta-catenin

Nuclear myosin VI was associated with cytoplasmic myosin VI (p=0.004) and

nuclear beta-catenin (p=0.009). E-cadherin and beta-catenin were associated in

the nucleus (p<0.001) and in the membranes and cytoplasm (p=0.02).

78

5.3.3 HuR and cyclooxygenase-2 (III)

HuR

There were 145 RCC tumours available to evaluate HuR expression. Nuclear

HuR immunopositivity was detected in 127 tumours (88%). Six (4%) of the

tumours were strongly positive, 31 (21%) were weakly positive and 108 (75%)

were negative for cytoplasmic HuR immunostaining. For further statistical

analysis, the subgroups of weakly and strongly positive cases were combined to

form a group of 37 positive tumours (25%), which was compared to the group of

negative cases.

Cytoplasmic HuR was associated with advanced stage (p=0.004), a higher T

class (p=0.04) and distant metastases (p=0.009). Tumours with cytoplasmic HuR

expression showed a tendency toward a higher Fuhrman grade (p=0.06). A

significant association was found between the histological subtype of RCC and

cytoplasmic HuR expression (p=0.01). Of the clear cell and chromophobic

carcinomas, 100 (78%) and 4 (80%) did not express HuR in the cytoplasm,

respectively, whereas 5 (56%) papillary carcinomas showed immunoreactivity for

cytoplasmic HuR. Both of the unclassified tumours expressed HuR. Cytoplasmic

HuR immunoexpression was more common in male than in female patients

(p=0.04). Nuclear HuR immunoreactivity was not associated with the

clinicopathological parameters studied. The immunoexpression of cytoplasmic

and nuclear HuR was associated (p=0.007).

The mean RCC-specific survival was longer for patients whose tumours did

not express cytoplasmic HuR (p=0.01). The mean RCC-specific survival times for

HuR-expressing and non-expressing tumours were 118 (95% CI 89–146) and 164

(95% CI 148–179) months, respectively. The lack of HuR expression was also a

prognostic factor of a more favourable RCC-specific survival in the subgroup of

clear cell RCCs (p=0.006). The mean RCC-specific survival was 108 (95% CI

76–140) and 162 (95% CI 146–178) months for cytoplasmic HuR

immunopositive and immunonegative clear cell RCCs, respectively. The HR for

HuR-expressing clear cell tumours was 2.45 (95% CI 1.27–4.75, p=0.008)

compared to clear cell carcinomas with no cytoplasmic HuR expression. In the

multivariate analysis adjusted for age, stage and grade, cytoplasmic HuR

expression was not found to be a prognostic factor. Nuclear HuR

immunoreactivity was not a prognostic factor in RCC-specific survival.

79

Cyclooxygenase-2

Out of 147 RCC tumours, cytoplasmic COX-2 immunostaining was weakly

positive in 22 (15%), moderately positive in 10 (7%) and strongly positive in 12

(8%) tumours. The negative and weakly positive cases and the moderately and

strongly positive cases were re-classified as negative (125 tumours, 85%) and

positive (22 tumours, 15%), respectively. No COX-2 immunostaining was

detected in the nucleus.

Cytoplasmic COX-2 expression was associated with advanced stage (p=0.04),

a higher T class (p=0.03), lymph node metastases (p=0.04), distant metastases

(p=0.02) and a higher nuclear grade (p=0.04). A relationship was found between

COX-2 expression and the histological type of RCC (p=0.007). Similar to

cytoplasmic HuR, COX-2 expression was more frequently detected in papillary (3

tumours, 33%) and unclassified (2 tumours, 100%) tumours than in clear cell (16

tumours, 12%) and chromophobic (one tumour, 20%) carcinomas.

Lack of COX-2 expression was a marker of better prognosis (p=0.02). The

mean RCC-specific survival times were 158 (95% CI 144–173) and 107 (95% CI

70–144) months for COX-2 immunonegative and immunopositive tumours,

respectively. Cytoplasmic COX-2 expression was a prognostic factor also in the

subgroup of tumours with clear cell histology. The mean RCC-specific survival

was 96 (95% CI 54–138) months for COX-2 immunopositive and 156 (95% CI

140–172) months for immunonegative clear cell RCCs (p=0.01). The HR for

COX-2-expressing clear cell carcinomas was 2.50 (95% CI 1.19–5.26, p=0.02).

In the analysis of five chromophobic carcinomas, cytoplasmic COX-2

immunopositivity was associated with unfavourable patient survival, with p=0.05.

In the Cox regression analysis for age, stage, grade and COX-2 expression, the

only statistically significant factor in RCC-specific survival was stage (p<0.001).

The relationship between HuR and cyclooxygenase-2

Cytoplasmic HuR was associated with cytoplasmic COX-2 (p<0.001). Both HuR

and COX-2 expression was present in 16 (11%) tumours whereas 102 (70%) of

the tumours were immunonegative for both of these markers. In addition,

cytoplasmic COX-2 expression and membranous E-cadherin were inversely

associated (p=0.002).

80

Cell culture experiments

In order to test the hypothesis that HuR regulates COX-2 expression in RCC cells,

human RCC cells were treated with HuR siRNA molecules. This treatment

suppressed HuR expression, as detected by immunoblotting. Furthermore, a

suppression of COX-2 protein expression after the HuR siRNA treatment was

observed. Control siRNA showed no effect on HuR or COX-2 expression. These

results indicated that HuR can regulate COX-2 expression in RCC cells.

5.3.4 Toll-like receptor 9 (IV)

Of the 138 RCC tumours available for the TLR9 immunoreactivity evaluation, 60

(44%) showed some nuclear immunopositivity. Twenty-one (15%) of the tumours

were strongly positive, 39 (28%) were moderately positive and 52 (38%) were

weakly positive for cytoplasmic TLR9 immunostaining. For further analyses, the

weakly, moderately and strongly positive cases were combined and re-classified

as TLR9-positive samples (112 tumours, 81%).

No correlation was found between the clinicopathological parameters studied

and TLR9 immunoexpression. The immunoexpression of TLR9 was observed in

100 (82%), 6 (67%), 4 (80%) and 2 (100%) tumours representing the histological

subtypes of clear cell RCC, papillary RCC, chromophobic RCC and unclassified

RCC, respectively. The expression of TLR9 and tumour necrosis was not

associated (data not shown).

In univariate analysis, RCC-specific survival was significantly longer for

patients whose tumours expressed cytoplasmic TLR9 compared to patients with

TLR9-immunonegative tumours (p=0.007). The mean RCC-specific survival time

was 112 (95% CI 76–147) months for tumours that did not express TLR9 in the

cytoplasm and 160 (95% CI 144–175) months for TLR9-positive tumours. The

expression of TLR9 was also a marker of more favourable RCC-specific survival

when the histological subtypes were analysed separately. The mean RCC-specific

survival period for cytoplasmic TLR9-expressing clear cell RCCs was 158

months (95% CI 142–175 months), whereas for immunonegative tumours it was

113 months (95% CI 74–152 months, p=0.02). The expression of TLR9 also

indicated a more favourable RCC-specific survival (p=0.05) in the analysis of

nine papillary RCCs. In the Cox regression analysis for cytoplasmic TLR9

expression, including age, stage and nuclear grade, TLR9 expression was an

independent prognostic factor in RCC-specific survival along with stage and

81

grade. The HR was 3.52 (95% CI 1.71–7.26, p=0.001) for tumours that did not

express TLR9 compared to TLR9-positive tumours. The expression of TLR9 was

also an independent marker of a more favourable RCC-specific prognosis in the

subgroup of clear cell RCCs. The HR was 3.59 (95% CI 1.65–7.80, p=0.001) for

TLR9-negative tumours compared to immunopositive clear cell RCCs. Nuclear

TLR9 was not a prognostic factor in RCC-specific survival.

5.3.5 Markers of the neuroendocrine system (V)

Out of 145 RCC tumours, 12 (8%) were immunopositive for serotonin. There

were 144 specimens available for the immunohistochemical evaluation of CD56,

of which 26 (18%) were immunopositive. Regarding immunopositivity for NSE,

69 cases out of 145 (48%) were detected. Synaptophysin immunopositivity was

found in 2 out of 148 tumours (1%). The immunoexpression of chromogranin A

was not observed in any of the 148 tumours (0%). The expression of serotonin

and CD56 (p<0.001), serotonin and synaptophysin (p=0.006) and CD56 and

synaptophysin (p=0.03) were associated.

The immunoexpression of NSE was more common in clear cell RCCs than in

the other subtypes (p=0.01). Other NE markers (serotonin, CD56 and

synaptophysin) were not found to be associated with histology or with the other

clinicopathological parameters, although the correlation between CD56

expression and a higher T class was almost significant (p=0.05).

In survival analysis, none of the neuroendocrine markers tested was found to

be a significant prognostic factor in RCC-specific survival, although serotonin-

immunopositive and CD56- and NSE-immunonegative tumours had a tendency

toward longer RCC-specific survival. The mean RCC-specific survival times for

serotonin-positive and negative tumours were 176 (95% CI 138–214) months and

148 (95% CI 133–163) months, respectively (p=0.3). The mean RCC-specific

survival was 156 (95% CI 140–172) months for CD56-immunonegative tumours

and 153 (95% 134–171) months for NSE-immunonegative tumours, whereas it

was 139 (95% CI 106–172) months and 149 (95% CI 127–171) months for

tumours that expressed CD56 and NSE, respectively (p=0.4 for CD56 and p=0.8

for NSE). Both patients with synaptophysin-immunopositive tumours were alive

at the end of follow-up. When the stainings (serotonin, CD56 and NSE) were

separately included in the Cox multivariate analysis with stage, Fuhrman grade

and age, the only independent prognostic factor for RCC-specific survival was

stage (p<0.001).

82

5.3.6 The prognostic accuracy of tested biomarkers

The most important prognostic factors of the tested biomarkers were cytoplasmic

TLR9, myosin VI and HuR, which improved the prognostic accuracy of the

multivariate survival analysis especially when used in combination. The most

important single prognostic factor was stage.

83

6 Discussion

In the search for novel prognostic markers for RCC we evaluated the expression

of 14 molecular markers by immunohistochemistry and examined their prognostic

significance. In addition, we explored the prognostic impact of widely used

clinicopathological parameters in our patient material.

6.1 Material and methodology

As our study population consisted of 77 females and 75 males, the gender

distribution differed from the normal male predominance (Chow et al. 2010),

although the predominance of females seemed to be coincidental. The commonly

recognized risk factors for RCC, i.e., smoking, hypertension and obesity (data not

shown) (Lipworth et al. 2006) were frequently found in our study population. The

majority of the tumours were operated on by radical nephrectomy, which was the

main type of operation performed for renal tumours in the 1990s. The clinical

practice at our hospital in the 1990s favoured preoperative embolization and it

was successfully performed for 24% of the patients. One half of the tumours were

asymptomatic and incidentally found by radiological examination, which is in

line with the results of previous studies (Pantuck et al. 2000). The distribution of

histological subtypes corresponds quite well with data presented in the literature,

although there was a slight predominance of clear cell carcinomas (88%) whereas

there were slightly fewer papillary and chromophobic carcinomas than previously

established (Kovács et al. 1997). The distribution of different nuclear grades was

not equal as there were only five grade I tumours whereas grade II tumours were

overrepresented. This minor discrepancy may be explained by the size of study

population. The stage distribution deviated a little from that generally reported as

almost one half of the tumours were stage I and only 12% of the patients had

metastatic disease (Motzer et al. 1996). However, each stage was represented and

thus this series can be considered unselected. During the follow-up period 29%

patients died of RCC, which is slightly less than the traditionally reported figure

of more than 40% (Landis et al. 1999, Bui et al. 2001).

There were many variables in each study that may have biased our results. It

is possible that some of the positive results were so-called false positives (alpha

error, type I error) due to coincidence, for example. On the other hand, as some of

the markers did not show any prognostic potential, there is a possibility of beta

84

errors (type II error), which may have been caused by the patient cohort being too

small, for example.

6.2 Patient survival and clinicopathological features as prognostic factors

The RCC-specific survival of whole study population was in line with the results

of previous reports (Tsui et al. 2000). Minor discrepancies in RCC-specific

survival could be explained by the size of the study population and slightly

different stage distributions. The prognostic significance of classic prognostic

factors is well described in the literature (Méjean et al. 2003, Volpe & Patard

2010) and the results from our study are in agreement with these previous

findings. In multivariate analysis the nuclear grade lost its prognostic significance,

which was also recognized in many previous studies (Delahunt et al. 2007).

The value of tumour-related anatomical parameters, including TNM stage,

stage grouping and tumour size, as well as time to appearance of metastasis, as

independent prognostic factors was confirmed in our study. Interestingly, large

localized tumours (T2, stage II) had a poorer prognosis in terms of shorter mean

survival and higher HR for death from RCC (under long-term surveillance) than

locally advanced RCCs (T3, stage III). This observation supports the results of a

previous study by Murphy and co-workers (Murphy et al. 2005), as well as

traditional clinical observations. However, the low number of cases in the

subgroups of pT2 and stage II reduces the reliability of this finding. The TNM

classification is a dynamic prognostic tool; its significance in predicting the

survival of RCC patients has been confirmed in multiple trials (Volpe & Patard

2010). Despite the revisions made over the years there is an ongoing debate

regarding further finetuning of the TNM staging concerning, for example, the

significance of tumour size, especially in locally advanced disease, sinus fat and

collecting system invasion, as well as infiltration into the vena cava wall. It might

be also considered whether or not there is a need to combine novel variables such

as performance status, symptoms or molecular profile in the TNM classification

system. (Lam et al. 2008, Delahunt 2009, Moch et al. 2009) For our material,

however, the TNM classification 2002 accurately predicted 5- and 10-year RCC-

specific survival.

The prognostic significance of several histological factors but not the

histological subtype or cystic structure was also confirmed in this study. The low

number of cases in the subgroups of papillary and chromophobic carcinomas and

85

the lack of an evaluation of the possible cystic structure in almost one half of the

cases may be the reasons why no statistically significant survival difference was

found in our study according to these variables. The undifferentiated carcinomas

showed a poor survival rate, as could be expected.

Although it is generally thought that incidentally found asymptomous

tumours have a better prognosis than symptomatic ones, not all studies, including

ours, support this (Mevorach et al. 1992, Jayson & Sanders 1998). Several earlier

studies indicated that the more favourable survival rates among asymptomatic

tumours is related to the smaller size and lower stage of these tumours (Pantuck et al. 2000, Méjean et al. 2003, Schips et al. 2003). In our material the incidentally

found and asymptomatic tumours did not differ in stage, T class or tumour size,

which presumably explains the similar survival rates observed. However, the

incidentally found tumours appeared to have a lower nuclear grade compared to

the symptomatic tumours (p=0.005), which has also been previously reported

(Pantuck et al. 2000).

6.3 Markers of the oxidative system (I)

Oxidative stress is known to promote carcinogenesis (reviewed in Klaunig et al. 1998) and oxidative stress and DNA damage have also been found in RCC

(Okamoto et al. 1994, Soini et al. 2006b). Thus, we decided to investigate some

of the markers involved in protection against oxidative stress in order to discover

whether or not they have any prognostic significance in RCC. The markers

selected were Nfr2 and Keap1, which work as a pair to sense and defend against

ROS (Motohashi & Yamamoto 2004), and Trx, which is an antioxidant enzyme

regulated by Nfr2 (Stacy et al. 2006).

6.3.1 Keap1 and Nfr2

The immunoexpression pattern of Nfr2 and Keap1 that was detected

corresponded to the pattern generally reported (McMahon et al. 2003, Itoh et al. 2004). The expression of Nfr2 was frequent in RCC whereas just over half of the

tumours were immunopositive for Keap1. The associations between these

markers reflected the earlier observations that Keap1 regulates the expression of

Nfr2 by binding it in the cytoplasm (Itoh et al. 2004) and that accumulated Nfr2

translocates from the cytoplasm into the nucleus (Kaspar et al. 2009).

86

The relationship between Keap1 expression and advanced stage largely

explains the poorer prognosis of Keap1-immunopositive tumours. In addition, the

higher grade RCC tumours were more frequently Keap1 positive. This is in

contrast to the situation in lung cancer, where reduced Keap1 expression and the

resulting nuclear accumulation of Nfr2 have been shown to stimulate cell growth

and to be related to poor survival rates (Ohta et al. 2008, Solis et al. 2010). It is

evident that the biological consequences of the Keap1-Nfr2 system have not been

thoroughly determined in cancer. By protecting DNA against oxidative and

xenobiotic damage and by modulating the redox balance, the activity of Nfr2 can

suppress cancer development (Satoh et al. 2010, Taguchi et al. 2011). Thus, it is

reasonable to suggest that the overexpression of Keap1 and the consequent down-

regulation of nuclear Nfr2 activity could promote carcinogenesis by repressing

defence mechanisms such as antioxidants against oxidative stress. Although

Keap1 and Nfr2 were associated in the cytoplasm, we could not, however, find

any significant association between Keap1 immunoexpression and nuclear Nfr2.

The expression of Nfr2 was not found to be a prognostic factor for RCC-specific

survival either. It is possible that the immunohistochemical quantification of

nuclear Nfr2 did not correspond to functionally active Nfr2. The modulation of

Nfr2 activity could be more complex than the ROS-induced modification of

cysteine residues of Keap1 alone. For example, the phosphorylation of serine of

Nfr2 can result in the release of Nfr2 from Keap1 and be another regulatory

mechanism for Nfr2 activity. (Apopa et al. 2008, Yamamoto et al. 2008, Kaspar et al. 2009) Nuclear Nrf2 expression is, however, an independent prognostic marker

in lung cancer (Merikallio et al. 2011). Consequently, the distribution of Nrf2 in

the nucleus may be cancer-type specific and relate to exposure to specific

carcinogens or specific treatments. Moreover, although the accumulation of Nfr2

and the consequently induced expression of cytoprotective genes have been

regarded as the main mechanisms of action of the Keap1-Nfr2 system, additional

pathways might also influence survival and contribute to cancer prognosis. Indeed,

there is evidence to show that Keap1 can, for example, regulate apoptosis

(Wakabayashi et al. 2010, Niture & Jaiswal 2011). As the prognostic significance

of Keap1 was not explained by Nfr2, these as yet unknown mechanisms might

explain the influence on survival that was observed in our study by Keap1.

87

6.3.2 Thioredoxin

The immunoexpression of Trx observed in our study was in line with that

previously reported for RCC (Soini et al. 2006b). Furthermore, nuclear

immunostaining for Trx has been previously reported in other cancers including

breast, cervical and gastric carcinomas (Fujii et al. 1991, Grogan et al. 2000,

Turunen et al. 2004). Thioredoxin is an Nfr2-regulated gene product and Nfr2

activation has been shown to result in the up-regulation of Trx (Stacy et al. 2006,

Singh et al. 2008), which was confirmed by the association between nuclear Nfr2

and Trx expression in our study.

Thioredoxin immunoreactivity was more frequently observed in well-

differentiated tumours than in poorly differentiated ones, and Trx-expressing

tumours were associated with longer survival. These observations are in

agreement with those of earlier studies that established the relationships between

various antioxidants and lower grade, more local stage and better survival in RCC

(Soini et al. 2006a, Soini et al. 2006b). It has been suggested that oxidative stress

and the formation of ROS could result in the induction of antioxidant enzymes

which, in turn, could protect cancer cells from further oxidative damage and

restrict additional genomic instability (Soini et al. 2006a, Soini et al. 2006b).

6.4 Myosin VI, E-cadherin and beta-catenin (II)

Invasion and metastasis are the hallmarks of cancer (reviewed in Hanahan &

Weinberg 2011). A cell adhesion molecule E-cadherin and a cytoplasmic

regulatory protein beta-catenin are both involved in these processes, as well as in

the pathogenesis of RCC (reviewed in Banumathy & Cairns 2010, van Kilsdonk et al. 2010). The previous observation that unconventional myosin VI is

associated with the E-cadherin-beta-catenin complex (Küssel-Andermann et al. 2000, Geisbrecht & Montell 2002) and has multiple roles that may contribute to

invasion and metastasis (Dunn et al. 2006, Knudsen 2006) inspired us to study the

immunoexpression of myosin VI in RCC tumours in an attempt to discover

whether or not myosin VI expression has any prognostic significance in RCC.

6.4.1 Myosin VI

The staining pattern of myosin VI was equal to patterns previously reported

(Vreugde et al. 2006, Sweeney & Houdusse 2007). In ovarian and prostate

88

carcinomas myosin VI expression was associated with a higher grade (Yoshida et al. 2004, Dunn et al. 2006) and thus our observation of a relationship between

cytoplasmic myosin VI immunoreactivity and greater differentiation is

controversial.

When confounding factors were eliminated in the survival analysis, myosin

VI expression was found to be an independent marker of poorer prognosis in

RCC. Moreover, myosin VI also predicted survival in the subgroup of low grade

tumours which is an interesting observation from a clinical point of view. Similar

observations of myosin VI as a marker of unfavourable cancer prognosis have

been reported in ovarian and prostate carcinomas and in leukaemia, where myosin

VI overexpression is related to adverse clinical features of tumour and cancer

progression. In detail, the knock-down of myosin VI inhibits cancer cell

spreading and colony formation. (Yoshida et al. 2004, Dunn et al. 2006, Jbireal et al. 2010)

Myosin VI is involved in cancer cell migration, which contributes to tumour

invasion in ovarian and prostate carcinomas and in leukaemia (Geisbrecht &

Montell 2002, Yoshida et al. 2004, Dunn et al. 2006, Buss & Kendrick-Jones

2008, Jbireal et al. 2010). The exact mechanism via which it enhances cell

migration is not fully understood. Myosin can produce a protrusive force by

binding to adhesion complexes such as E-cadherin-beta-catenin in the membrane

and, by pushing the actin filaments outwards, it can move the cell forward (Buss

& Kendrick-Jones 2008). Myosin VI might also be involved in the internalization

and recycling of key mediators of cell migration (Yoshida et al. 2004) and

increased myosin VI expression could facilitate tumour invasion by enhancing

cancer cells to respond to growth factor stimulation (Knudsen 2006). During

tumour progression changes occur in cell polarity, proteins at cell-cell junctions

become reorganized and, furthermore, myosin VI translocates from cell-cell

junctions to membrane ruffles, which could be related to both invasion and

metastasis (Knudsen 2006). A recent study on prostate cancer found that the

overexpression of myosin VI enhances the secretion of VEGF, which could also

be related to cancer progression (Puri et al. 2010). These mechanisms could

explain the prognostic impact of myosin expression observed in our study.

Vascular endothelial growth factor (VEGF) plays a central role in the

pathogenesis of RCC (reviewed in Pfaffenroth & Linehan 2008) and thus the

relationship between myosin VI and VEGF (Puri et al. 2010) is intriguing and

warrants further investigation.

89

6.4.2 E-cadherin

The membranous immunoexpression of E-cadherin detected in the present study

was slightly less frequent than commonly reported for RCC (Terpe et al. 1993,

Katagiri et al. 1995, Shimazui et al. 1997). Altered E-cadherin expression is,

however, a less frequent phenomenon in RCC compared to other epithelial

carcinomas, presumably due to the origin of RCC. The majority of RCCs

originate from proximal tubular epithelial cells, which normally do not express E-

cadherin at all. Instead, E-cadherin is present in the distal tubules and collecting

ducts of the normal kidney. (Katagiri et al. 1995) Our study replicated the

previous observations of the association between histology and E-cadherin

expression in RCC (Gervais et al. 2007).

Some previous studies associated the down-regulation of E-cadherin with

advanced stage (Katagiri et al. 1995, Shimazui et al. 1997) and an unfavourable

clinical outcome in RCC (Katagiri et al. 1995). These results were not replicated

in our study. However, the prognostic role of E-cadherin was not proven in every

previous study (Shimazui et al. 1997).

In addition to membranous expression, nuclear E-cadherin immunostaining

has been reported in some cancers (Gervais et al. 2007, Chetty & Serra 2008).

The exact mechanism by which E-cadherin translocates to the nucleus is unclear,

but it is assumed that it is related to some cooperative molecules such as beta-

catenin (Chetty & Serra 2008). In a previous study by Gervais and co-workers,

nuclear E-cadherin expression was correlated with a low Fuhrman grade, VHL

immunoreactivity and better prognosis, and it was only observed in clear cell

variants (Gervais et al. 2007). Although nuclear E-cadherin expression was also

detected in our study, these observations could not be statistically confirmed,

which was presumably related to methodological differences and a different study

population.

6.4.3 Beta-catenin

The normal immunoexpression pattern of beta-catenin is membranous and/or

cytoplasmic (Kim et al. 2000). Beta-catenin gene mutations, which seem to be

uncommon in the pathogenesis of RCC (Kim et al. 2000), could result in nuclear

and/or cytoplasmic accumulation of beta-catenin (Bilim et al. 2000). Previous

reports of beta-catenin immunoexpression in RCC are variable, as both increased

and decreased beta-catenin immunoreactivity have been observed. For example,

90

in the study of Shimazui and co-workers, 24% of RCCs showed abnormal, i.e.

absent, or heterogeneous immunoreactivity for beta-catenin (Shimazui et al. 1997). In other studies, decreased membranous or cytoplasmic immunoreactivity

was observed in 16% and 42% of RCC tumours, respectively (Bilim et al. 2000,

Aaltomaa et al. 2004). Some studies established a connection between the normal

immunostaining pattern or cytoplasmic accumulation of beta-catenin and clear

cell carcinoma (Shimazui et al. 1997 Kim et al. 2000, Guo et al. 2001). The

comparison of individual studies seems to be quite infeasible due to fundamental

differences in immunohistochemical procedures and evaluations. However, it

seems that cytoplasmic beta-catenin expression was less usual and nuclear

immunoreactivity more frequent in our study population than reported previously

(Aaltomaa et al. 2004). Previous reports of the association between beta-catenin

and tumour differentiation are controversial: on the one hand an association

between preserved beta-catenin expression and better differentiation was reported

(Bilim et al. 2000), whereas on the other hand the down-regulation of

membranous beta-catenin was reportedly associated with low grade clear cell

carcinomas (Guo et al. 2001). We observed an association between nuclear beta-

catenin expression and better differentiation, whereas cytoplasmic expression was

not correlated with nuclear grade.

The prognostic influence of beta-catenin differs between various cancers.

Beta-catenin expression has been connected to invasion and reduced survival in

colon and liver cancers, for example, whereas it has been found to be related to a

more favourable prognosis in malignant melanoma and transitional cell

carcinoma (Shimazui et al. 1996, Arozarena et al. 2011). Studies concerning the

prognostic value of beta-catenin in RCC have mainly reported that decreased

beta-catenin expression is related to advanced stage, tumour recurrence and

decreased patient survival (Shimazui et al. 1997, Bilim et al. 2000, Kim et al. 2000, Zhu et al. 2000, Kuroiwa et al. 2001, Aaltomaa et al. 2004). However, it is

very difficult to compare these previous studies as they have mixed and differing

immunostaining patterns and the cut-off values for decreased, normal and

increased levels of immunoexpression were unique for each study, among other

methodological differences. Our study found that cytoplasmic beta-catenin

expression was related to poorer RCC-specific survival in clear cell carcinoma. It

is evident that beta-catenin plays convoluted roles in the pathogenesis of RCC and

thus the impact of beta-catenin immunoexpression might vary in this malignancy.

91

6.4.4 Associations between myosin VI, E-cadherin and beta-catenin

We could not prove an association between myosin VI and E-cadherin-beta-

catenin in the cytoplasm, unlike previously described (Küssel-Andermann et al. 2000, Geisbrecht & Montell 2002). However, there was an association between

nuclear myosin VI and beta-catenin. As myosin VI is a cytoplasmic protein

(Sweeney & Houdusse 2007), its role in the nucleus remains obscure, but it did

not appear to have any prognostic significance in RCC. Nuclear beta-catenin

expression was previously reported in various malignancies such as colon, gastric

and endometrial carcinomas (Iwamoto et al. 2000, Ougolkov et al. 2000, Saegusa et al. 2001), as well as in RCC (Aaltomaa et al. 2004). We suggest that myosin VI

and beta-catenin could somehow co-operate or interact. The role of beta-catenin

as a regulator of transcription factors has been clearly described (Behrens et al. 1996, Karim et al. 2004). In fact, myosin VI has also been linked to the regulation

of gene expression and transcriptional modulation (Dunn et al. 2006, Vreugde et al. 2006).

The positive relationship between E-cadherin and beta-catenin expression has

previously been shown in gastrointestinal carcinomas (Han et al. 2000, Gervais et al. 2007, Serra et al. 2007) and it was also confirmed in our study. Moreover, E-

cadherin binding has been shown to prevent beta-catenin nuclear localization

(Orsulic et al. 1999). Indeed, we also found that nuclear beta-catenin was more

frequently absent in clear cell RCC with membranous E-cadherin expression (data

not shown). It is assumed that nuclear E-cadherin is a cleavage product of the

membranous full-length E-cadherin and a direct association was previously found

between membranous and nuclear E-cadherin expression in RCC (Gervais et al. 2007). This association was also confirmed in our study.

6.5 HuR and cyclooxygenase-2 (III)

Inflammation is regarded as one of the risk factors for cancer development and

progression (reviewed in Farrow & Evers 2002). Up-regulation of the

inflammation-associated enzyme COX-2 has been shown to be correlated with

adverse clinicopathological features and more aggressive tumour behaviour in

various malignomas including RCC (Sheehan et al. 1999, Miyata et al. 2003). In

this study, we decided to find out whether or not COX-2 expression is regulated

by the mRNA stability protein HuR in RCC and whether or not HuR expression

has some prognostic value in RCC.

92

6.5.1 HuR

The expression of HuR has been found to be greater and more frequently

cytoplasmic in RCC than in the normal kidney (López de Silanes et al. 2003). The

nuclear immunoexpression of HuR observed in the present study is in line with

the results of a previous study by Danilin et al. (Danilin et al. 2010). However,

cytoplasmic HuR reactivity seemed to be less frequent in our study population

than in earlier studies (López de Silanes et al. 2003, Danilin et al. 2010). This

discrepancy could be explained, at least partly, by methodological aspects since

we used paraffin-embedded, formalin-fixed specimens compared to the freshly

harvested tumour samples used in the study by Danilin and co-workers, and by

the low number of cases in the study by López de Silanes et al. We detected a significant association between HuR and COX-2 expression at

the protein level and showed the regulation of COX-2 expression by HuR in RCC

cell lines, which were previously shown in several other malignomas (Dixon et al. 2001, Erkinheimo et al. 2003, Denkert et al. 2004b, Mrena et al. 2005, Denkert et al. 2006, Lim et al. 2007). The similar relationship observed in the present study

between cytoplasmic HuR expression and clinicopathological parameters of

advanced stage and decreased cancer survival was previously demonstrated in

other carcinomas (Erkinheimo et al. 2003, Denkert et al. 2004a, Denkert et al. 2004b, Heinonen et al. 2005, Mrena et al. 2005, Denkert et al. 2006, Lim et al. 2007).

Although the ability to induce COX-2 expression is regarded as the

explanation for the prognostic significance of HuR expression in many cancers

(Erkinheimo et al. 2003, Mrena et al. 2005), HuR also regulates the stability

and/or translation of numerous other mRNAs involved in malignant

transformation and cancer promotion (López de Silanes et al. 2005, Hinman &

Lou 2008,). The HuR protein enhances angiogenesis as it increases HIF-1α and

VEGF expression, which was previously shown in colon carcinoma cells, for

example (Levy et al. 1998, Dixon et al. 2001, Sheflin et al. 2004). The

overexpression of HuR was shown to be related to increased cell proliferation by

the stabilization of mRNAs of epidermal growth factor (EGF) (Sheflin et al. 2004), granulocyte macrophage colony-stimulating factor (GM-CSF) (López de

Silanes et al. 2005) and proto-oncogenic and proliferative factors such as c-myc,

c-fos, cyclin A and cyclin B1 (Peng et al. 1998, Wang et al. 2000, Wang et al. 2001). Furthermore, HuR inhibits apoptosis via the antiapoptotic protein

prothymosin α (ProTα) (López de Silanes et al. 2005). Other HuR-regulated

93

target proteins have been reported in studies on colon carcinoma, for example

beta-catenin (López de Silanes et al. 2003), as well as some invasion and

metastasis related proteins including matrix metalloproteinase 9 (MMP-9),

urokinase plasminogen activator (uPA) and the metastasis-associated protein

(López de Silanes et al. 2005). In addition, there is evidence to show that HuR

induces the expression of immunosuppressive cytokines, which may help the

tumour to avoid immune surveillance (López de Silanes et al. 2005). Many

immunomodulating factors such as TGF-β, TNF-α and interleukins IL-6 and IL-8

have been found to be the target mRNAs of HuR (Dixon et al. 2001, Nabors et al. 2001).

Evidence has shown that HuR has a central role in the pathogenesis of RCC

(Danilin et al. 2010), where HuR was found to be overexpressed in all of the

samples of conventional RCC tested, regardless of the VHL status (Danilin et al. 2010). Renal cell carcinoma cell culture specimens showed that the down-

regulation of HuR reduces proliferation (Danilin et al. 2009) and facilitates

apoptosis (Danilin et al. 2010). Moreover, HuR knock-down was found to inhibit

RCC tumour growth in both cell cultures and an in vivo animal model (Danilin et al. 2010). In conventional RCC, HuR expression has shown to be associated for

example with VEGF and HIF-2α expression as well as with the constitutive

activation of the Pl3K/AKT pathway (Danilin et al. 2010).

The relationship between HuR and VHL has been shown in several studies on

RCC. The stabilization of tumour suppressor p53 mRNA by HuR is pVHL-

dependent and could partly explain the tumour suppressive functions of pVHL

that have been observed in RCC (Galbán et al. 2003). On the other hand, pVHL

reduces HuR expression via sequestration, resulting in decreased expression of

type I insulin-like growth factor receptor (IGF1R) in clear cell RCC (Yuen et al. 2007). In turn, IGF1R is one of the key mediators of tumour cell survival and its

expression is connected to a worse prognosis in clear cell RCC (Parker et al. 2002). In addition, the parathyroid hormone-related protein (PTHrP), which is

responsible for humoural hypercalcaemia associated with RCC, has growth

factor-like activities and is an independent prognostic factor of patient outcome in

RCC (Iwamura et al. 1999), where it was shown to be regulated by HuR in a

VHL-dependent manner (Danilin et al. 2009).

Although nuclear HuR expression has been shown to correlate with

clinicopathological features, including stage and grade, and patient prognosis in

gynaecological carcinomas (Lim et al. 2007, Yi et al. 2009), these observations

were not replicated in our RCC material.

94

Based on the results of our study, HuR regulates COX-2 expression in RCC,

which presumably explains the prognostic influence of HuR in this disease.

However, HuR can also be connected to the pathogenesis of RCC via several

other routes, as described above. The HuR protein has been suggested as a

potential target in cancer therapy due to its presumed central role in

carcinogenesis (López de Silanes et al. 2005, Lim et al. 2007). The antitumoural

activity of miRNAs repressing HuR expression has been shown in breast

carcinoma in vitro and in cervical carcinoma in vivo (Guo et al. 2009,

Abdelmohsen et al. 2010). The therapeutic role of HuR inhibitors or antagonists

should also be evaluated in RCC.

6.5.2 Cyclooxygenase-2

The expression of COX-2 is a common feature in RCC and is reported in over

one half of RCC tumours (Miyata et al. 2003, Hashimoto et al. 2004, Tuna et al. 2004, Mungan et al. 2006). However, in our study population COX-2 expression

was found less frequently than in previous studies. Part of this discrepancy could

be explained by the cut-off values used, for example we classified weakly

positive tumours as negative tumours, and partly by differences in the

immunohistochemical procedures and study population. Our results were,

however, in line with those of a previous Finnish study where COX-2

immunostaining was reported in 26% of primary RCC tumours (Kankuri-

Tammilehto et al. 2010). The cytoplasmic expression pattern observed in our

study was similar to previously reported patterns (Miyata et al. 2003, Tuna et al. 2004). The expression of COX-2 was less frequent in clear cell carcinomas than

in other histological subtypes. Previously, Hashimoto et al. found that COX-2

expression is more common in the granular cell subtype than in clear cell variants

(Hashimoto et al. 2004). As the classification “granular” is no longer

recommended, it was not documented in our study protocol.

In our material, the expression of COX-2 was associated with

clinicopathological variables related to advanced disease and reduced RCC-

specific survival. The relationships observed between COX-2 immunoexpression

and various clinicopathological features such as TNM stage, tumour size and

nuclear grade were also observed in several previous studies on RCC (Miyata et al. 2003, Hashimoto et al. 2004, Tuna et al. 2004, Rini et al. 2006), although not

in every study (Yoshimura et al. 2004, Cho et al. 2005, Kankuri-Tammilehto et al. 2010). In light of the results from previous studies, the prognostic potential of

95

COX-2 seems to be contradictory in RCC. Our material appeared to show that

COX-2 immunoreactivity is not an independent prognostic factor for RCC-

specific survival, as found in a previous study by Miyata et al. (Miyata et al. 2003). Several other studies were not able to prove a prognostic significance of

COX-2 expression (Tuna et al. 2004, Cho et al. 2005), although an association

was reported between COX-2 expression and more favourable overall survival in

patients with metastatic RCC treated with immunotherapy (Kankuri-Tammilehto et al. 2010). However, a comparison of these earlier studies is unreliable because

of the different study populations and study protocols used, as well as variable

immunohistochemical procedures and cut-off values. To the best of our

knowledge, our study population was the largest so far for this type of

immunohistochemical study and our follow-up time was considerably longer than

in most previous studies.

We observed an inverse relationship between COX-2 and membranous E-

cadherin expression, which could reflect the previous observation that the down-

regulation of COX-2 alters cell adhesion molecules such as E-cadherin and

inhibits invasion in RCC (Chen et al. 2004a). Thus, this finding of an association

between COX-2 and E-cadherin expression warrants further investigation. It has

been hypothesized that beta-catenin could activate COX-2 expression through

transcription factor LEF-1 (Prescott & White 1996). In colorectal carcinoma, APC

mutations have been shown to be correlated with both the up-regulation of COX-

2 and increased nuclear beta-catenin expression (Dimberg et al. 2001). In our

study, however, no association was found between COX-2 and beta-catenin

expression.

6.6 Toll-like receptor 9 (IV)

The idea of studying TLR9 expression and its relationship to patient survival

arose from previous observations in other types of cancer where TLR9 expression

has been shown to be related to higher grades and the invasiveness of cancer

(Ilvesaro et al. 2007, Jukkola-Vuorinen et al. 2008, Väisänen et al. 2010).

The cytoplasmic expression pattern of TLR9 observed in the present study

corresponds to expression patterns reported in other cancers (Jukkola-Vuorinen et al. 2008, Väisänen et al. 2010). The nuclear TLR9 expression observed in the

present study was not reported previously and might represent unspecific staining.

The expression of TLR9 was related to a more favourable clinical outcome in

RCC. Furthermore, it was not associated with the clinicopathological parameters

96

and, based on the results of our study, TLR9 expression is indeed an independent

prognostic factor for RCC-specific survival. Although there were a few previous

studies on the prognostic value of TLR9 expression in cancer, these studies

mainly connected TLR9 with adverse clininopathological features of cancer such

as poor differentiation (Jukkola-Vuorinen et al. 2008, Berger et al. 2010,

Väisänen et al. 2010, Wang et al. 2010). Moreover, the mechanism by which

TLR9 stimulation promotes cancer cell invasion through the up-regulation of

MMP activity in brain, breast, lung and prostate cancers in vitro has also been

determined (Merrell et al. 2006, Ilvesaro et al. 2007, Ren et al. 2007). In addition,

the stimulation of TLR9 has been shown to promote cell proliferation, inhibit

apoptosis and regulate several genes and transcription factors related to

tumorigenesis and cancer progression in hepatocellular carcinoma (Tanaka et al. 2010). In glioma, TLR9 expression is associated with poor patient outcome

(Wang et al. 2010).

It is evident that the function of TLR9 expression differs in RCC from that of

other cancers as the lack of TLR9 confers aggressive behaviour to renal

carcinoma cells. However, some parallel observations have been reported in other

malignomas. In recurrent breast carcinoma, TLR9 expression in fibroblast-like

cells has been shown to be correlated with a low probability of metastasis

(González-Reyes et al. 2010) and, in neuroblastoma, TLR9 expression is

inversely associated with disease stage (Brignole et al. 2010). Although TLR9

expression has been connected with higher nuclear grade and although TLR9

stimulation has been shown to induce cancer invasion in prostate carcinoma, its

activation by immunomodulatory nucleotides was reported to induce apoptosis in vitro and to cause the regression of tumour growth in an in vivo xenograft model

of prostate cancer (Rayburn et al. 2006). Thus, the contribution of either high or

low TLR9 expression to the pathophysiology of cancer may be highly tumour

specific.

The immunogenic nature of RCC is generally recognized. As cytokines such

as IFN-α and IL-2 can produce complete and durable responses in metastatic

RCC (Vogelzang et al. 1992, Finley et al. 2011), one might hypothesize that the

immune response generated by ligand binding to TLR9 is the mechanism by

which TLR9 expression results in more favourable survival in RCC. Upon

recognition of the CpG-DNA ligand, TLR9 recruits specific intracellular adaptor

proteins to initiate signalling pathways that activate transcription factors such as

NF-kappaB. The eventual outcome is an immune reaction characterized by the

stimulation of B-cell proliferation, the activation of macrophages and dendritic

97

cells and the production of inflammatory mediators such as interferon,

interleukins and other inflammatory cytokines. (Akira & Hemmi 2003, Wagner

2004, Kawai & Akira 2008) We suggested that in the absence of TLR9 expression

such immune responses are not evoked in RCC cells and that cancer cells are less

susceptible to immunosurveillance and can progress further. Further studies are

needed to clarify the mechanism and the role of TLR9 expression in RCC. The

underlying yet hypothetical mechanisms of TLR9 activation should be examined

in the future.

Hypoxia and the compensatory hyperactivation of angiogenesis are important

features in the pathogenesis of RCC tumours (reviewed in Banumathy & Cairns

2010). Hypoxia has been shown to increase the expression of various TLRs

including TLR9 (Kuhlicke et al. 2007, Liu et al. 2009). Moreover, HIF-1 has been

shown to coordinate the induction of TLRs (Kuhlicke et al. 2007). Although in

our study population TLR9 expression and tumour necrosis were not found to be

associated, it would be interesting to discover whether or not hypoxia and HIF-1

regulate TLR9 expression in RCC cancer cells.

The prognostic role of TLR9 evokes expectations of the therapeutic potential

of TLR9 agonists in RCC. Synthetic TLR9 agonists are being examined as anti-

tumour immunomodulants against various cancers and they seem to have

potential as anticancer treatments in preclinical and early clinical trials (Krieg

2008). However, a synthetic TLR9 agonist only showed modest antitumour

activity in advanced RCC (Thompson et al. 2009). It would be reasonable to

study whether or not there is difference in the response rates between TLR9-

expressing and non-expressing tumours.

6.7 Markers of neuroendocrine activity (V)

Neuroendocrine cells are involved in the regulation of cell growth and

differentiation and neuroendocrine products may stimulate cell proliferation and

tumour growth (Abrahamsson 1999, Bostwick et al. 2002, Erickson & Lloyd

2004). In some cancers, neuroendocrine differentiation has been shown to have

some prognostic impact (Syversen et al. 1995, Abrahamsson 1999). These

observations were the rationale for studying neuroendocrine activity as a

prognostic factor for RCC.

Serotonin immunoexpression is absent in metastatic RCC (Edgren et al. 1996), which was confirmed in our study population (data not shown). In light of

the results of previous studies, the prognostic impact of serotonin seems to be

98

contradictory in cancer. In prostate cancer, both increased and decreased serotonin

immunoexpression have contributed to advanced or aggressive disease (Bostwick et al. 2002, Dizeyi et al. 2004). On the one hand, by reducing tumour blood flow,

presumably through its vasoconstrictive effect, serotonin inhibits tumour growth

(Vicaut et al. 2000). On the other hand, serotonin is a growth factor either directly

or through synergy with other growth factors and it stimulates DNA synthesis and

cell proliferation and differentiation (Vicaut et al. 2000, Siddiqui et al. 2005). The

tumour growth promoting impact of serotonin has been established in several

malignancies (Vicaut et al. 2000) and 5HT antagonists, in turn, inhibit growth and

cell division in a variety of tumours (Tutton & Barkla 1987, Siddiqui et al. 2005).

In high-grade prostate carcinoma, serotonin stimulates cancer cells and serotonin

receptors are overexpressed. In prostate cancer cell lines serotonin activates

PI3K/Akt signalling pathways, for example. (Dizeyi et al. 2004, Dizeyi et al. 2011) However, the tumour-promoting effect seems to be dose dependent. High

doses of serotonin may have a direct mitogenic effect on tumour cells whereas

low doses may reduce tumour growth (Vicaut et al. 2000). Suppressed plasma

serotonin levels have been detected in metastatic RCC (Jungwirth et al. 2008), as

well as a correlation between elevated serotonin plasma levels and longer survival

(Edgren et al. 1996). In a way, our observation of longer RCC-specific survival

being related to serotonin immunoexpression supports these previous reports. It is

possible that our population was too small and serotonin expression too rare to

show statistically significant differences in the survival analysis. However, based

on our results and on previous observations of metastatic RCC, the role of

serotonin in RCC seems to be more of a suppressive one than a promotional one.

We confirmed a previous observation of CD56 expression in less than 20% of

RCCs (Daniel et al. 2003). There are some observations in the literature linking

CD56 expression to unfavourable clinical outcomes of cancer, for example

metastasis (Koukoulis et al. 1998, Daniel et al. 2000, Daniel et al. 2001). In RCC,

CD56 immunopositivity is associated with advanced stage, aggressive Fuhrman

grade and poor survival rates (Daniel et al. 2003). In our study population the

relationship between CD56 expression and a higher T class was almost significant

(p=0.05) and the survival analysis suggested an unfavourable RCC-specific

prognosis in CD56-immunopositive tumours.

Neurone-specific enolase is the most sensitive and unspecific neuroendocrine

marker. Based on the results of previous studies, NSE expression is extremely

frequent in RCC which may indicate that NSE is produced by the RCC tumour

itself (Haimoto et al. 1986, Edgren et al. 1996, Rasmuson et al. 1999). The

99

discrepancy in NSE expression observed in the present study can be explained, at

least partly, by the use of formalin-fixed, paraffin-embedded tumour samples

compared to cryostat sections (Haimoto et al. 1986) and by a different stage

distribution and larger study population (Edgren et al. 1996). In addition, these

previous studies reported staining patterns other than the pure cytoplasmic

staining found in our study. Nevertheless, NSE immunopositivity was more

common in metastatic than localized RCC in our material, as in previous studies,

although the difference was not statistically significant (data not shown). In RCC,

an increase in serum NSE is a common phenomenon which is associated with

advanced disease and a poor patient outcome. A decline in serum NSE can be

observed after the treatment of RCC. (Haimoto et al. 1986, Takashi et al. 1989,

Yaman et al. 1996, Rasmuson et al. 1999) As epithelial cells of the proximal

tubules where RCC originates do not normally express NSE, NSE expression is

believed to occur during renal oncogenesis (Haimoto et al. 1986). Although NSE

expression was somewhat related to decreased RCC-specific survival in our study

population, the immunoexpression of NSE does not seem to be a statistically

significant prognostic factor in RCC based on our material.

We were not able to detect any chromogranin A immunoexpression, which is

in line with the results of previous reports where absent or extremely rare

chromogranin A expression was reported in RCC (Edgren et al. 1996, Rasmuson et al. 1999). Increased serum chromogranin A levels have been found in RCC

patients, which presumably reflects sources of chromogranin A other than RCC

tumours (Edgren et al. 1996, Rasmuson et al. 1999).

According to the results of our study specific neuroendocrine differentiation

is rare in RCC since there were only two synaptophysin positive tumours (1%) in

our material. To the best of our knowledge, this was the first study to detect the

immunoexpression of synaptophysin in RCC. In a small previous study on

metastatic RCC, no immunoreactivity was found for synaptophysin at all (Edgren et al. 1996). The synaptophysin-expressing tumours in the present study had

TNM classifications of T1aN0M0 and T3aN0M0 and excellent RCC-specific

survival as the patients were alive after 10 and 17 years of follow-up, respectively.

There are no reports in the literature of an association between synaptophysin

expression and more favourable cancer survival. Instead, synaptophysin

immunoreactiviy is associated with a higher tumour grade in breast carcinoma

(Makretsov et al. 2003). In undifferentiated endometrial carcinoma,

synaptophysin expression did not seem to have any prognostic value (Taraif et al.

100

2009). It would be fascinating to discover whether this indolent clinical behaviour

is related to the neuroendocrine differentiation observed in these RCC tumours.

6.8 Challenges in searching for prognostic biomarkers for renal cell carcinoma

There are several factors that could explain the difficulties encountered in

attempting to discover prognostic biomarkers suitable for clinical practice in RCC.

The studies in this field are mostly single institutional retrospective studies with a

limited number of patients and a relatively short follow-up period (Nogueira &

Kim 2008, Volpe & Patard 2010). Compared to previous studies with a similar

research framework and objective, the advantages of our study were its long and

detailed follow-up period and the rather large study population. Due to the

retrospective nature of our study not all anamnestic and clinical data were

available afterwards, which may have biased our results, particularly concerning

the prognostic value of preoperative symptoms and signs, as well as some of the

laboratory findings. The style of documentation was more undefined in the 1990s

than it is nowadays and, in some cases, collecting all of the data we required was

laborious since several different sources had to be checked. However, the most

relevant data were available and each patient was appropriately staged. In

addition, the follow-up was completed in every case. Some discrepancies between

our reports and previous studies can be explained by the different patient cohorts.

In our material, stage and grade distribution deviated slightly from generally

reported. Although the number of patients was rather large compared to other

studies, the study cohort might still have been too small to distinguish minor

statistical differences. However, the patient cohort was large enough to point out

the most relevant prognostic factors, such as stage.

A fundamental problem in this kind of study is the lack of recognition of

distinct histological subtypes of RCC in the analysis (Eichelberg et al. 2009).

Although each histological subtype was analysed separately in our study, we

primarily analysed the whole population mainly because of the low number of

cases of each of the histological subtypes other than clear cell carcinoma. In our

study population the results from the separate survival analysis were non-

conclusive. The molecular markers we used are “generic” in carcinogenesis

(Eichelberg et al. 2009), which justified studying them in the whole RCC

population as well. According to the separate analyses performed, the biomarkers

studied seemed to be suitable for prognostic factors regardless of the histological

101

subtype. In the future, study populations must be larger if separate analyses for

papillary, chromophobic and collecting duct carcinomas are to be performed.

The traditional method used in this kind of study for investigating prognostic

tissue markers is the immunohistochemistry of archival tumour material, which

addresses certain concerns (Di Napoli & Signoretti 2009). The results of

immunohistochemical studies can be influenced by numerous parameters from

conditions during surgery such as anaesthesia, blood pressure variations, blood

loss and renal artery clumping to variables in tissue handling, including

preparation in the pathology department, such as room temperature, tissue drying

and oxidation, fixation-related variables including the type of fixative, its volume

and concentration, as well as specimen size, and variables during storage such as

temperature and length (Di Napoli & Signoretti 2009, Eichelberg et al. 2009). In

the immunohistochemical staining procedure there are numerous variables that

are not standardized and can influence the immunoexpression of proteins of

interest, such as the antibody used and its dilution, pretreatment and staining

techniques (Chetty & Serra 2008, Volpe & Patard 2010). In archival tumour

material the type and duration of fixation can influence antigen preservation and

recognition by a specific antibody. Protein immunoexpression has been shown to

be higher in fresh tumour material than in archival formalin-fixed, paraffin-

embedded tissue samples. (Abrahamsson et al. 1987, Di Napoli & Signoretti 2009)

Our stainings were performed in the paraffin-embedded and formalin-fixed

archival material. As some previous studies used freshly harvested tumour

specimens the results are not fully comparable. Antigen retrieval by heating and

enzymatic treatment such as that used in the present study partly reverse the

formalin-fixation effect (Di Napoli & Signoretti 2009). Although the maximum

storage length was more than 20 years, the immunoexpression of the different

markers studied was generally quite well preserved as the results were in line with

those of previous studies. One concern regarding our study material was

preoperative embolization, which was performed for one fourth of the patients

and possibly somehow influenced the immunoexpression of the markers studied.

However, the tumours were selected in the multitissue block from a vital, well-

preserved tumour area.

Renal cell carcinomas are often large, necrotic and heterogeneous tumours

and therefore the selection of the most representative tumour area for

immunohistochemical study is not simple (Di Napoli & Signoretti 2009). The

preparation of multitissue blocks demands patience, skill and experience. Due to

technical aspects, tumour necrosis or cut-off of the tumour area, for example,

102

some samples are always lost during immunohistochemical procedures (sampling

error). When multiple multitissue block series are prepared the tumour area is not

identical in each slide. In the present study, some samples were excluded from the

analysis of each series because of the above-mentioned reasons.

The evaluation and quantification of staining results can be important pitfalls

in the immunohistochemical procedure. When an evaluation is performed by eye,

it is not particularly reproducible or objective. Different scoring procedures lack

standardization. (Di Napoli & Signoretti 2009, Volpe & Patard 2010) In addition,

immunohistochemical assessments of proteins do not necessarily correlate with

their biological activity (Frank et al. 2000). These sources of error also concern

this study.

The lack of reproducibility was a fundamental problem among all of the

studies that tried to discover a prognostic biomarker for RCC (Volpe & Patard

2010). Although one succeeded in finding an independent prognostic factor, the

others were not able to replicate the results or observations were contradictory.

Many of these above-mentioned problems can be solved by better study designs.

There is a need for large, prospective, multicenter studies in which different

histological types are separated (Eichelberg et al. 2009). The

immunohistochemical procedures used should be standardized so that others can

replicate the procedure and externally validate the results (Crispen et al. 2008,

Lam et al. 2008). Although there is much criticism against retrospective, single

institutional immunohistochemical studies they provide important basic

information about potential prognostic tissue markers that could later be

externally validated in a larger series.

6.9 Future aspects

Based on our results, the expression of Keap1, myosin VI, HuR and TLR9 should

be evaluated in other RCC cohorts to assess their reproducibility and significance

as prognostic markers in RCC. The future goal is to develop a prognostic tool that

combines traditional prognostic factors such as stage and grade with the

molecular and genetic profile of RCC tumours to provide prognostic information

that is not available with classical prognostic factors alone or with current

nomograms. We suggest that TLR9, HuR and myosin VI may be promising

candidates for novel nomograms since, particularly when combined, they

improved the prognostic accuracy of the multivariate survival analysis. Ongoing

research recognizes the enormous amount of potential biomarkers, which first

103

have to be evaluated at a basic level in immunohistochemical studies such as ours,

and then the most promising markers can be further re-evaluated and validated in

a larger, prospective series. The biomarkers evaluated thus far are primarily based

on their expression in tumour tissue; in the future, however, we should attempt to

find markers that can also be used in blood or urine samples so that such analyses

would be suitable for daily clinical practice, and also be cost-effective. The

discovery of “PSA for RCC” needs diligent work and, among other things, a

certain amount of luck, too.

104

105

7 Summary and conclusions

The traditional prognostic factors are not sufficient to accurately predict the

clinical outcome of each RCC patient and there are no prognostic biomarkers for

RCC suitable for daily clinical use. As we want to treat our patients in the best

way possible, a more precise prognostication of individual patients than is

currently possible is necessary. In this study, novel tissue markers were examined

and evaluated as prognostic markers for RCC.

The main findings of the current study with 152 RCC patients are as follows:

1. The immunoexpression of Keap1 is related to advanced stage and poorer

disease-specific survival in RCC.

2. Cytoplasmic myosin VI immunoexpression is an independent marker of

reduced RCC-specific survival and is associated with lower nuclear grade.

Nuclear myosin VI and beta-catenin are associated.

3. Cyclooxygenase-2 expression is regulated by HuR in RCC. Cytoplasmic HuR

expression is associated with reduced RCC-specific survival.

4. The immunoexpression of TLR9 is common in RCC and does not associate

with stage or grade. The lack of TLR9 expression predicts short RCC-specific

survival and TLR9 expression is an independent prognostic biomarker in RCC.

5. Although neuroendocrine activity is detectable in RCC, the markers studied, i.e.

serotonin, CD56, NSE, chromogranin A and synaptophysin, are not potential

prognostic markers in RCC.

In conclusion, several potential prognostic tissue markers for RCC were

identified in this study. In particular, cytoplasmic TLR9 and myosin VI, which

were acknowledged as independent prognostic factors for RCC-specific survival,

should be further investigated in large, prospective series and be validated

externally. As RCC is an immunogenic malignancy, the more favourable

prognosis associated with TLR9 expression is intriguing and warrants further

investigation. The underlying, presumably immunologic responses should be

determined. In addition, it should be discovered whether or not TLR9 expression

predicts the therapeutic response of TLR9 agonist treatment in advanced RCC.

Myosin VI predicted the outcome of RCC-patient also in the subgroup of well-

differentiated tumours, which are thought to have a favourable prognosis, making

it an interesting marker from a clinical point of view. The central role of HuR in

carcinogenesis and its prognostic significance we observed in RCC are also

fascinating and may contribute to the therapeutic aspects of RCC.

106

So far, the prognosis of an individual RCC patient is mainly determined by

staging in clinical practice. Although we introduced several potential biomarkers

that could help, especially when combined, to predict the clinical outcome of

RCC, based on the results of the present study cancer stage is still the most

important single prognostic factor for RCC. Thus, all proposed novel prognostic

factors should be compared to stage and proven to possess more predictive

accuracy than staging.

107

References

Aaltomaa S, Lipponen P, Karja V, Lundstedt S, Lappi J & Kosma VM (2004) The expression and prognostic value of alpha-, beta- and gamma-catenins in renal cell carcinoma. Anticancer Res 24(4): 2407–2413.

Abbate M, Brown D & Bonventre JV (1999) Expression of NCAM recapitulates tubulogenic development in kidneys recovering from acute ischemia. Am J Physiol 277(3 Pt 2): F454–463.

Abdelmohsen K, Kim MM, Srikantan S, Mercken EM, Brennan SE, Wilson GM, Cabo R & Gorospe M (2010) miR-519 suppresses tumor growth by reducing HuR levels. Cell Cycle 9(7): 1354–1359.

Aberle H, Schwartz H & Kemler R (1996) Cadherin-catenin complex: protein interactions and their implications for cadherin function. J Cell Biochem 61(4): 514–523.

Abrahamsson PA (1999) Neuroendocrine cells in tumour growth of the prostate. Endocr Relat Cancer 6(4): 503–519.

Abrahamsson PA, Falkmer S, Fält K & Grimelius L (1989) The course of neuroendocrine differentiation in prostatic carcinomas. An immunohistochemical study testing chromogranin A as an “endocrine marker”. Pathol Res Pract 185(3): 373–380.

Abrahamsson PA, Wadström LB, Alumets J, Falkmer S & Grimelius L (1987) Peptide-hormone- and serotonin-immunoreactive tumour cells in carcinoma of the prostate. Pathol Res Pract 182(3): 298–307.

Akira S & Hemmi H (2003) Recognition of pathogen-associated molecular patterns by TLR family. Immunol Lett 85(2): 85–95.

Akira S, Takeda K & Kaisho T (2001) Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immunol 2(8): 675–680.

Anderson KV, Jürgens G & Nüsslein-Volhard C (1985) Establishment of dorsal-ventral polarity in the Drosophila embryo: genetic studies on the role of the Toll gene product. Cell 42(3): 779–789.

Apopa PL, He X & Ma Q (2008) Phosphorylation of Nrf2 in the transcription activation domain by casein kinase 2 (CK2) is critical for the nuclear translocation and transcription activation function of Nrf2 in IMR-32 neuroblastoma cells. J Biochem Mol Toxicol 22(1): 63–76.

Arden SD, Puri C, Au JS, Kendrick-Jones J & Buss F (2007) Myosin VI is required for targeted membrane transport during cytokinesis. Mol Biol Cell 18(12): 4750–4761.

Ariyoshi Y, Kato K, Ishiguro Y, Ota K, Sato T & Suchi T (1983) Neuron-specific enolase as a new tumor marker. Gan To Kagaku Ryoho 10(8): 1744–1753.

Arozarena I, Bischof H, Gilby D, Belloni B, Dummer R & Wellbrock C (2011) In melanoma, beta-catenin is a suppressor of invasion. Oncogene 30(45): 4531–4543.

Awakura Y, Nakamura E, Ito N, Kamoto T & Ogawa O (2008) Methylation-associated silencing of SFRP1 in renal cell carcinoma. Oncol Rep 20(5): 1257–1263.

Baiyee D & Banner B (2006) Mismatch repair genes in renal cortical neoplasms. Hum Pathol 37(2): 185–189.

108

Baker A, Payne CM, Briehl MM & Powis G (1997) Thioredoxin, a gene found overexpressed in human cancer, inhibits apoptosis in vitro and in vivo. Cancer Res 57(22): 5162–5167.

Banumathy G & Cairns P (2010) Signaling pathways in renal cell carcinoma. Cancer Biol Ther 10(7): 658–664.

Battagli C, Uzzo RG, Dulaimi E, Ibáñez de Cáceres I, Krassenstein R, Al-Saleem T, Greenberg RE & Cairns P (2003) Promoter hypermethylation of tumor suppressor genes in urine from kidney cancer patients. Cancer Res 63(24): 8695–8699.

Becker F, Siemer S, Hack M, Humke U, Ziegler M & Stöckle M (2006) Excellent long-term cancer control with elective nephron-sparing surgery for selected renal cell carcinomas measuring more than 4 cm. Eur Urol 49(6): 1058–1063; discussion 1063–1064.

Becker KF, Kremmer E, Eulitz M, Becker I, Handschuh G, Schuhmacher C, Müller W, Gabbert HE, Ochiai A, Hirohashi S & Höfler H (1999) Analysis of E-cadherin in diffuse-type gastric cancer using a mutation-specific monoclonal antibody. Am J Pathol 155(6): 1803–1809.

Behrens J, von Kries JP, Kühl M, Bruhn L, Wedlich D, Grosschedl R & Birchmeier W (1996) Functional interaction of beta-catenin with the transcription factor LEF-1. Nature 382(6592): 638–642.

Bensalah K, Montorsi F & Shariat SF (2007) Challenges of cancer biomarker profiling. Eur Urol 52(6): 1601–1609.

Berger R, Fiegl H, Goebel G, Obexer P, Ausserlechner M, Doppler W, Hauser-Kronberger C, Reitsamer R, Egle D, Reimer D, Müller-Holzner E, Jones A & Widschwendter M (2010) Toll-like receptor 9 expression in breast and ovarian cancer is associated with poorly differentiated tumors. Cancer Sci 101(4): 1059–1066.

Berggren M, Gallegos A, Gasdaska JR, Gasdaska PY, Warneke J & Powis G (1996) Thioredoxin and thioredoxin reductase gene expression in human tumors and cell lines, and the effects of serum stimulation and hypoxia. Anticancer Res 16(6B): 3459–3466.

Biemesderfer D, Mentone SA, Mooseker M & Hasson T (2002) Expression of myosin VI within the early endocytic pathway in adult and developing proximal tubules. Am J Physiol Renal Physiol 282(5): F785–94.

Bilim V, Kawasaki T, Katagiri A, Wakatsuki S, Takahashi K & Tomita Y (2000) Altered expression of beta-catenin in renal cell cancer and transitional cell cancer with the absence of beta-catenin gene mutations. Clin Cancer Res 6(2): 460–466.

Blaschko H, Comline RS, Schneider FH, Silver M & Smith AD (1967) Secretion of a chromaffin granule protein, chromogranin, from the adrenal gland after splanchnic stimulation. Nature 215(5096): 58–59.

Bluemke K, Bilkenroth U, Meye A, Fuessel S, Lautenschlaeger C, Goebel S, Melchior A, Heynemann H, Fornara P & Taubert H (2009) Detection of circulating tumor cells in peripheral blood of patients with renal cell carcinoma correlates with prognosis. Cancer Epidemiol Biomarkers Prev 18(8): 2190–2194.

109

Bostwick DG, Qian J, Pacelli A, Zincke H, Blute M, Bergstralh EJ, Slezak JM & Cheng L (2002) Neuroendocrine expression in node positive prostate cancer: correlation with systemic progression and patient survival. J Urol 168(3): 1204–1211.

Brennan CM & Steitz JA (2001) HuR and mRNA stability. Cell Mol Life Sci 58(2): 266–277.

Brignole C, Marimpietri D, Di Paolo D, Perri P, Morandi F, Pastorino F, Zorzoli A, Pagnan G, Loi M, Caffa I, Erminio G, Haupt R, Gambini C, Pistoia V & Ponzoni M (2010) Therapeutic targeting of TLR9 inhibits cell growth and induces apoptosis in neuroblastoma. Cancer Res 70(23): 9816–9826.

Bui MH, Seligson D, Han KR, Pantuck AJ, Dorey FJ, Huang Y, Horvath S, Leibovich BC, Chopra S, Liao SY, Stanbridge E, Lerman MI, Palotie A, Figlin RA & Belldegrun AS (2003) Carbonic anhydrase IX is an independent predictor of survival in advanced renal clear cell carcinoma: implications for prognosis and therapy. Clin Cancer Res 9(2): 802–811.

Bui MH, Visapää H, Seligson D, Kim H, Han KR, Huang Y, Horvath S, Stanbridge EJ, Palotie A, Figlin RA & Belldegrun AS (2004) Prognostic value of carbonic anhydrase IX and KI67 as predictors of survival for renal clear cell carcinoma. J Urol 171(6 Pt 1): 2461–2466.

Bui MH, Zisman A, Pantuck AJ, Han KR, Wieder J & Belldegrun AS (2001) Prognostic factors and molecular markers for renal cell carcinoma. Expert Rev Anticancer Ther 1(4): 565–575.

Buskens CJ, Van Rees BP, Sivula A, Reitsma JB, Haglund C, Bosma PJ, Offerhaus GJ, Van Lanschot JJ & Ristimäki A (2002) Prognostic significance of elevated cyclooxygenase 2 expression in patients with adenocarcinoma of the esophagus. Gastroenterology 122(7): 1800–1807.

Buss F, Arden SD, Lindsay M, Luzio JP & Kendrick-Jones J (2001) Myosin VI isoform localized to clathrin-coated vesicles with a role in clathrin-mediated endocytosis. EMBO J 20(14): 3676–3684.

Buss F & Kendrick-Jones J (2008) How are the cellular functions of myosin VI regulated within the cell? Biochem Biophys Res Commun 369(1): 165–175.

Buss F, Spudich G & Kendrick-Jones J (2004) Myosin VI: cellular functions and motor properties. Annu Rev Cell Dev Biol 20: 649–676.

Campbell SC, Novick AC & Bukowski RM (2007) Renal tumors. In: Wein AJ, Kavoussi LR, Novick AC, Partin AW & Peters CA (eds) Campbell-Walsh urology, 9th edition. Philadelphia, Saunders Elsevier: 1582, 1591–1598.

Cao Y & Prescott SM (2002) Many actions of cyclooxygenase-2 in cellular dynamics and in cancer. J Cell Physiol 190(3): 279–286.

Chance B, Sies H & Boveris A (1979) Hydroperoxide metabolism in mammalian organs. Physiol Rev 59(3): 527–605.

Chen Q, Shinohara N, Abe T, Harabayashi T & Nonomura K (2004a) Impact of cyclooxygenase-2 gene expression on tumor invasiveness in a human renal cell carcinoma cell line. J Urol 172(6 Pt 1): 2153–2157.

110

Chen Q, Shinohara N, Abe T, Watanabe T, Nonomura K & Koyanagi T (2004b) Significance of COX-2 expression in human renal cell carcinoma cell lines. Int J Cancer 108(6): 825–832.

Chetty R & Serra S (2008) Nuclear E-cadherin immunoexpression: from biology to potential applications in diagnostic pathology. Adv Anat Pathol 15(4): 234–240.

Chitalia VC, Foy RL, Bachschmid MM, Zeng L, Panchenko MV, Zhou MI, Bharti A, Seldin DC, Lecker SH, Dominguez I & Cohen HT (2008) Jade-1 inhibits Wnt signalling by ubiquitylating beta-catenin and mediates Wnt pathway inhibition by pVHL. Nat Cell Biol 10(10): 1208–1216.

Cho DS, Joo HJ, Oh DK, Kang JH, Kim YS, Lee KB & Kim SJ (2005) Cyclooxygenase-2 and p53 expression as prognostic indicators in conventional renal cell carcinoma. Yonsei Med J 46(1): 133–140.

Chow WH, Dong LM & Devesa SS (2010) Epidemiology and risk factors for kidney cancer. Nat Rev Urol 7(5): 245–257.

Crispen PL, Boorjian SA, Lohse CM, Leibovich BC & Kwon ED (2008) Predicting disease progression after nephrectomy for localized renal cell carcinoma: the utility of prognostic models and molecular biomarkers. Cancer 113(3): 450–460.

Dahabreh J, Stathopoulos GP, Koutantos J & Rigatos S (2009) Lung carcinoid tumor biology: treatment and survival. Oncol Rep 21(3): 757–760.

Daniel L, Bouvier C, Chetaille B, Gouvernet J, Luccioni A, Rossi D, Lechevallier E, Muracciole X, Coulange C & Figarella-Branger D (2003) Neural cell adhesion molecule expression in renal cell carcinomas: relation to metastatic behavior. Hum Pathol 34(6): 528–532.

Daniel L, Durbec P, Gautherot E, Rouvier E, Rougon G & Figarella-Branger D (2001) A nude mice model of human rhabdomyosarcoma lung metastases for evaluating the role of polysialic acids in the metastatic process. Oncogene 20(8): 997–1004.

Daniel L, Trouillas J, Renaud W, Chevallier P, Gouvernet J, Rougon G & Figarella-Branger D (2000) Polysialylated-neural cell adhesion molecule expression in rat pituitary transplantable tumors (spontaneous mammotropic transplantable tumor in Wistar-Furth rats) is related to growth rate and malignancy. Cancer Res 60(1): 80–85.

Danilin S, Sourbier C, Thomas L, Lindner V, Rothhut S, Dormoy V, Helwig JJ, Jacqmin D, Lang H & Massfelder T (2010) Role of the RNA-binding protein HuR in human renal cell carcinoma. Carcinogenesis 31(6): 1018–1026.

Danilin S, Sourbier C, Thomas L, Rothhut S, Lindner V, Helwig JJ, Jacqmin D, Lang H & Massfelder T (2009) von Hippel-Lindau tumor suppressor gene-dependent mRNA stabilization of the survival factor parathyroid hormone-related protein in human renal cell carcinoma by the RNA-binding protein HuR. Carcinogenesis 30(3): 387–396.

de Reijke TM, Bellmunt J, van Poppel H, Marreaud S & Aapro M (2009) EORTC-GU group expert opinion on metastatic renal cell cancer. Eur J Cancer 45(5): 765–773.

Delahunt B (2009) Advances and controversies in grading and staging of renal cell carcinoma. Mod Pathol 22(Suppl 2): S24–36.

111

Delahunt B, Bethwaite PB & Nacey JN (2007) Outcome prediction for renal cell carcinoma: evaluation of prognostic factors for tumours divided according to histological subtype. Pathology 39(5): 459–465.

Delahunt B, Bethwaite PB, Nacey JN & Ribas JL (1993) Proliferating cell nuclear antigen (PCNA) expression as a prognostic indicator for renal cell carcinoma: comparison with tumour grade, mitotic index, and silver-staining nucleolar organizer region numbers. J Pathol 170(4): 471–477.

Delahunt B, Eble JN, McCredie MR, Bethwaite PB, Stewart JH & Bilous AM (2001) Morphologic typing of papillary renal cell carcinoma: comparison of growth kinetics and patient survival in 66 cases. Hum Pathol 32(6): 590–595.

Denkert C, Koch I, von Keyserlingk N, Noske A, Niesporek S, Dietel M & Weichert W (2006) Expression of the ELAV-like protein HuR in human colon cancer: association with tumor stage and cyclooxygenase-2. Mod Pathol 19(9): 1261–1269.

Denkert C, Weichert W, Pest S, Koch I, Licht D, Köbel M, Reles A, Sehouli J, Dietel M & Hauptmann S (2004a) Overexpression of the embryonic-lethal abnormal vision-like protein HuR in ovarian carcinoma is a prognostic factor and is associated with increased cyclooxygenase 2 expression. Cancer Res 64(1): 189–195.

Denkert C, Weichert W, Winzer KJ, Müller BM, Noske A, Niesporek S, Kristiansen G, Guski H, Dietel M & Hauptmann S (2004b) Expression of the ELAV-like protein HuR is associated with higher tumor grade and increased cyclooxygenase-2 expression in human breast carcinoma. Clin Cancer Res 10(16): 5580–5586.

DeWitt DL & Smith WL (1988) Primary structure of prostaglandin G/H synthase from sheep vesicular gland determined from the complementary DNA sequence. Proc Natl Acad Sci U S A 85(5): 1412–1416.

Di Napoli A & Signoretti S (2009) Tissue biomarkers in renal cell carcinoma: issues and solutions. Cancer 115(Suppl 10): 2290–2297.

Díaz-Flores L, Gutiérrez R & Varela H (1994) Angiogenesis: an update. Histol Histopathol 9(4): 807–843.

DiBiase SJ, Valicenti RK, Schultz D, Xie Y, Gomella LG & Corn BW (1997) Palliative irradiation for focally symptomatic metastatic renal cell carcinoma: support for dose escalation based on a biological model. J Urol 158(3 Pt 1): 746–749.

Dimberg J, Hugander A, Sirsjö A & Söderkvist P (2001) Enhanced expression of cyclooxygenase-2 and nuclear beta-catenin are related to mutations in the APC gene in human colorectal cancer. Anticancer Res 21(2A): 911–915.

Dixon DA, Tolley ND, King PH, Nabors LB, McIntyre TM, Zimmerman GA & Prescott SM (2001) Altered expression of the mRNA stability factor HuR promotes cyclooxygenase-2 expression in colon cancer cells. J Clin Invest 108(11): 1657–1665.

Dizeyi N, Bjartell A, Nilsson E, Hansson J, Gadaleanu V, Cross N & Abrahamsson PA (2004) Expression of serotonin receptors and role of serotonin in human prostate cancer tissue and cell lines. Prostate 59(3): 328–336.

Dizeyi N, Hedlund P, Bjartell A, Tinzl M, Austild-Taskén K & Abrahamsson PA (2011) Serotonin activates MAP kinase and PI3K/Akt signaling pathways in prostate cancer cell lines. Urol Oncol 29(4): 436–445.

112

Doki Y, Shiozaki H, Tahara H, Inoue M, Oka H, Iihara K, Kadowaki T, Takeichi M & Mori T (1993) Correlation between E-cadherin expression and invasiveness in vitro in a human esophageal cancer cell line. Cancer Res 53(14): 3421–3426.

Donskov F & von der Maase H (2006) Impact of immune parameters on long-term survival in metastatic renal cell carcinoma. J Clin Oncol 24(13): 1997–2005.

Dorudi S, Sheffield JP, Poulsom R, Northover JM & Hart IR (1993) E-cadherin expression in colorectal cancer. An immunocytochemical and in situ hybridization study. Am J Pathol 142(4): 981–986.

Droemann D, Albrecht D, Gerdes J, Ulmer AJ, Branscheid D, Vollmer E, Dalhoff K, Zabel P & Goldmann T (2005) Human lung cancer cells express functionally active Toll-like receptor 9. Respir Res 6: 1.

Dubois RN, Abramson SB, Crofford L, Gupta RA, Simon LS, Van De Putte LB & Lipsky PE (1998) Cyclooxygenase in biology and disease. FASEB J 12(12): 1063–1073.

Dulaimi E, Ibáñez de Cáceres I, Uzzo RG, Al-Saleem T, Greenberg RE, Polascik TJ, Babb JS, Grizzle WE & Cairns P (2004) Promoter hypermethylation profile of kidney cancer. Clin Cancer Res 10(12 Pt 1): 3972–3979.

Dunn TA, Chen S, Faith DA, Hicks JL, Platz EA, Chen Y, Ewing CM, Sauvageot J, Isaacs WB, De Marzo AM & Luo J (2006) A novel role of myosin VI in human prostate cancer. Am J Pathol 169(5): 1843–1854.

Eberhart CE, Coffey RJ, Radhika A, Giardiello FM, Ferrenbach S & DuBois RN (1994) Up-regulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology 107(4): 1183–1188.

Eble JN, Togashi K, Pisani P, Merino MJ, Eccles DM, Linehan WM, Algaba F, Zbar B, Kovács G, Kleihues P, Geurts van Kessel A, Kiuru M, Launonen V, Herva R, Aaltonen LA, Neumann HPH, Pavlovich GP, Grignon DJ, Bonsib SM, Moch H, Delahunt B, Störkel S, Martignoni G, van der Berg E, Srigley JR, Davis CJ (2004) Tumours of the Kidney. In: Eble JN, Sauter GS, Epstein JI & Sesterhenn IA (eds) World Health Organization Classification of tumours. Pathology and genetics of tumours of the urinary system and male genital organs. Lyon, IARCPress: 9–36, 43.

Edgren M, Stridsberg M, Kalknar KM & Nilsson S (1996) Neuroendocrine markers; chromogranin, pancreastatin and serotonin in the management of patients with advanced renal cell carcinoma. Anticancer Res 16(6B): 3871–3874.

Eichelberg C, Junker K, Ljungberg B & Moch H (2009) Diagnostic and prognostic molecular markers for renal cell carcinoma: a critical appraisal of the current state of research and clinical applicability. Eur Urol 55(4): 851–863.

Eling TE, Thompson DC, Foureman GL, Curtis JF & Hughes MF (1990) Prostaglandin H synthase and xenobiotic oxidation. Annu Rev Pharmacol Toxicol 30: 1–45.

Erickson LA & Lloyd RV (2004) Practical markers used in the diagnosis of endocrine tumors. Adv Anat Pathol 11(4): 175–189.

Erkinheimo T-L, Lassus H, Sivula A, Sengupta S, Furneaux H, Hla T, Haglund C, Butzow R & Ristimäki A (2003) Cytoplasmic HuR expression correlates with poor outcome and with cyclooxygenase 2 expression in serous ovarian carcinoma. Cancer Res 63(22): 7591–7594.

113

Esteban MA, Tran MG, Harten SK, Hill P, Castellanos MC, Chandra A, Raval R, O'brien TS & Maxwell PH (2006) Regulation of E-cadherin expression by VHL and hypoxia-inducible factor. Cancer Res 66(7): 3567–3575.

Evans AJ, Russell RC, Roche O, Burry TN, Fish JE, Chow VW, Kim WY, Saravanan A, Maynard MA, Gervais ML, Sufan RI, Roberts AM, Wilson LA, Betten M, Vandewalle C, Berx G, Marsden PA, Irwin MS, Teh BT, Jewett MA & Ohh M (2007) VHL promotes E2 box-dependent E-cadherin transcription by HIF-mediated regulation of SIP1 and snail. Mol Cell Biol 27(1): 157–169.

Fan XC & Steitz JA (1998) Overexpression of HuR, a nuclear-cytoplasmic shuttling protein, increases the in vivo stability of ARE-containing mRNAs. EMBO J 17(12): 3448–3460.

Farrow B & Evers BM (2002) Inflammation and the development of pancreatic cancer. Surg Oncol 10(4): 153–169.

Finley DS, Pantuck AJ & Belldegrun AS (2011) Tumor biology and prognostic factors in renal cell carcinoma. Oncologist 16(Suppl 2): 4–13.

Finnish Cancer Registry (2011) Cancer statistics. URI: http://www.cancer.fi/ syoparekisteri/en/statistics/cancer-statistics/. Cited 2011/9/4.

Flanigan RC, Mickisch G, Sylvester R, Tangen C, Van Poppel H & Crawford ED (2004) Cytoreductive nephrectomy in patients with metastatic renal cancer: a combined analysis. J Urol 171(3): 1071–1076.

Fletcher L, Rider CC & Taylor CB (1976) Enolase isoenzymes. III. Chromatographic and immunological characteristics of rat brain enolase. Biochim Biophys Acta 452(1): 245–252.

Frank I, Blute ML, Cheville JC, Lohse CM, Weaver AL & Zincke H (2002) An outcome prediction model for patients with clear cell renal cell carcinoma treated with radical nephrectomy based on tumor stage, size, grade and necrosis: the SSIGN score. J Urol 168(6): 2395–2400.

Frank J, Pompella A & Biesalski HK (2000) Histochemical visualization of oxidant stress. Free Radic Biol Med 29(11): 1096–1105.

Freemont AJ & Hoyland JA (1996) Cell adhesion molecules. Clin Mol Pathol 49(6): M321–30.

Fuhrman SA, Lasky LC & Limas C (1982) Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 6(7): 655–663.

Fujii S, Nanbu Y, Nonogaki H, Konishi I, Mori T, Masutani H & Yodoi J (1991) Coexpression of adult T-cell leukemia-derived factor, a human thioredoxin homologue, and human papillomavirus DNA in neoplastic cervical squamous epithelium. Cancer 68(7): 1583–1591.

Galbán S, Martindale JL, Mazan-Mamczarz K, López de Silanes I, Fan J, Wang W, Decker J & Gorospe M (2003) Influence of the RNA-binding protein HuR in pVHL-regulated p53 expression in renal carcinoma cells. Mol Cell Biol 23(20): 7083–7095.

Gamallo C, Palacios J, Suárez A, Pizarro A, Navarro P, Quintanilla M & Cano A (1993) Correlation of E-cadherin expression with differentiation grade and histological type in breast carcinoma. Am J Pathol 142(4): 987–993.

114

Garin-Chesa P, Fellinger EJ, Huvos AG, Beresford HR, Melamed MR, Triche TJ & Rettig WJ (1991) Immunohistochemical analysis of neural cell adhesion molecules. Differential expression in small round cell tumors of childhood and adolescence. Am J Pathol 139(2): 275–286.

Gasdaska JR, Berggren M & Powis G (1995) Cell growth stimulation by the redox protein thioredoxin occurs by a novel helper mechanism. Cell Growth Differ 6(12): 1643–1650.

Gasdaska PY, Oblong JE, Cotgreave IA & Powis G (1994) The predicted amino acid sequence of human thioredoxin is identical to that of the autocrine growth factor human adult T-cell derived factor (ADF): thioredoxin mRNA is elevated in some human tumors. Biochim Biophys Acta 1218(3): 292–296.

Gazdar AF, Helman LJ, Israel MA, Russell EK, Linnoila RI, Mulshine JL, Schuller HM & Park JG (1988) Expression of neuroendocrine cell markers L-dopa decarboxylase, chromogranin A, and dense core granules in human tumors of endocrine and nonendocrine origin. Cancer Res 48(14): 4078–4082.

Geiger C, Nössner E, Frankenberger B, Falk CS, Pohla H & Schendel DJ (2009) Harnessing innate and adaptive immunity for adoptive cell therapy of renal cell carcinoma. J Mol Med 87(6): 595–612.

Geisbrecht ER & Montell DJ (2002) Myosin VI is required for E-cadherin-mediated border cell migration. Nat Cell Biol 4(8): 616–620.

Gervais ML, Henry PC, Saravanan A, Burry TN, Gallie BL, Jewett MA, Hill RP, Evans AJ & Ohh M (2007) Nuclear E-cadherin and VHL immunoreactivity are prognostic indicators of clear-cell renal cell carcinoma. Lab Invest 87(12): 1252–1264.

Gill IS, Remer EM, Hasan WA, Strzempkowski B, Spaliviero M, Steinberg AP, Kaouk JH, Desai MM & Novick AC (2005) Renal cryoablation: outcome at 3 years. J Urol 173(6): 1903–1907.

Giroldi LA, Bringuier PP & Schalken JA (1994) Defective E-cadherin function in urological cancers: clinical implications and molecular mechanisms. Invasion Metastasis 14(1–6): 71–81.

Gnarra JR, Tory K, Weng Y, Schmidt L, Wei MH, Li H, Latif F, Liu S, Chen F & Duh FM (1994) Mutations of the VHL tumour suppressor gene in renal carcinoma. Nat Genet 7(1): 85–90.

González-Reyes S, Marín L, González L, González LO, del Casar JM, Lamelas ML, González-Quintana JM & Vizoso FJ (2010) Study of TLR3, TLR4 and TLR9 in breast carcinomas and their association with metastasis. BMC Cancer 10: 665.

Grogan TM, Fenoglio-Prieser C, Zeheb R, Bellamy W, Frutiger Y, Vela E, Stemmerman G, Macdonald J, Richter L, Gallegos A & Powis G (2000) Thioredoxin, a putative oncogene product, is overexpressed in gastric carcinoma and associated with increased proliferation and increased cell survival. Hum Pathol 31(4): 475–481.

Gumbiner BM (1995) Signal transduction of beta-catenin. Curr Opin Cell Biol 7(5): 634–640.

115

Guo L, Kuroda N, Miyazaki E, Hayashi Y, Toi M, Naruse K, Hiroi M, Ashida S, Shuin T & Enzan H (2001) The complementary role of beta-catenin in diagnosing various subtypes of renal cell carcinomas and its up-regulation in conventional renal cell carcinomas with high nuclear grades. Oncol Rep 8(3): 521–526.

Guo X, Wu Y & Hartley RS (2009) MicroRNA-125a represses cell growth by targeting HuR in breast cancer. RNA Biol 6(5): 575–583.

Haimoto H, Takashi M, Koshikawa T, Asai J & Kato K (1986) Enolase isozymes in renal tubules and renal cell carcinoma. Am J Pathol 124(3): 488–495.

Han AC, Soler AP, Tang CK, Knudsen KA & Salazar H (2000) Nuclear localization of E-cadherin expression in Merkel cell carcinoma. Arch Pathol Lab Med 124(8): 1147–1151.

Hanahan D & Weinberg RA (2011) Hallmarks of cancer: the next generation. Cell 144(5): 646–674.

Hashimoto Y, Kondo Y, Kimura G, Matsuzawa I, Sato S, Ishizaki M, Imura N, Akimoto M & Hara S (2004) Cyclooxygenase-2 expression and relationship to tumour progression in human renal cell carcinoma. Histopathology 44(4): 353–359.

Haven CJ, Wong FK, van Dam EW, van der Juijt R, van Asperen C, Jansen J, Rosenberg C, de Wit M, Roijers J, Hoppener J, Lips CJ, Larsson C, Teh BT & Morreau H (2000) A genotypic and histopathological study of a large Dutch kindred with hyperparathyroidism-jaw tumor syndrome. J Clin Endocrinol Metab 85(4): 1449–1454.

Heidenreich A, Ravery V & European Society of Oncological Urology (2004) Preoperative imaging in renal cell cancer. World J Urol 22(5): 307–315.

Heinonen M, Bono P, Narko K, Chang SH, Lundin J, Joensuu H, Furneaux H, Hla T, Haglund C & Ristimäki A (2005) Cytoplasmic HuR expression is a prognostic factor in invasive ductal breast carcinoma. Cancer Res 65(6): 2157–2161.

Herman JG, Latif F, Weng Y, Lerman MI, Zbar B, Liu S, Samid D, Duan DS, Gnarra JR & Linehan WM (1994) Silencing of the VHL tumor-suppressor gene by DNA methylation in renal carcinoma. Proc Natl Acad Sci U S A 91(21): 9700–9704.

Hernberg M, Turunen JP, Muhonen T & Pyrhönen S (1997) Tumor-infiltrating lymphocytes in patients with metastatic melanoma receiving chemoimmunotherapy. J Immunother 20(6): 488–495.

Herrem CJ, Tatsumi T, Olson KS, Shirai K, Finke JH, Bukowski RM, Zhou M, Richmond AL, Derweesh I, Kinch MS & Storkus WJ (2005) Expression of EphA2 is prognostic of disease-free interval and overall survival in surgically treated patients with renal cell carcinoma. Clin Cancer Res 11(1): 226–231.

Hinman MN & Lou H (2008) Diverse molecular functions of Hu proteins. Cell Mol Life Sci 65(20): 3168–3181.

Hinshaw JL & Lee FT Jr (2004) Image-guided ablation of renal cell carcinoma. Magn Reson Imaging Clin N Am 12(3): 429–47.

116

Hirata H, Hinoda Y, Nakajima K, Kawamoto K, Kikuno N, Kawakami K, Yamamura S, Ueno K, Majid S, Saini S, Ishii N & Dahiya R (2009) Wnt antagonist gene DKK2 is epigenetically silenced and inhibits renal cancer progression through apoptotic and cell cycle pathways. Clin Cancer Res 15(18): 5678–5687.

Hofstädter F, Knüchel R & Rüschoff J (1995) Cell proliferation assessment in oncology. Virchows Arch 427(3): 323–341.

Huang M, Stolina M, Sharma S, Mao JT, Zhu L, Miller PW, Wollman J, Herschman H & Dubinett SM (1998) Non-small cell lung cancer cyclooxygenase-2-dependent regulation of cytokine balance in lymphocytes and macrophages: up-regulation of interleukin 10 and down-regulation of interleukin 12 production. Cancer Res 58(6): 1208–1216.

Huang WC, Elkin EB, Levey AS, Jang TL & Russo P (2009) Partial nephrectomy versus radical nephrectomy in patients with small renal tumors--is there a difference in mortality and cardiovascular outcomes? J Urol 181(1): 55–61; discussion 61–62.

Hutson TE (2011) Targeted therapies for the treatment of metastatic renal cell carcinoma: clinical evidence. Oncologist 16(Suppl 2): 14–22.

Ilvesaro JM, Merrell MA, Swain TM, Davidson J, Zayzafoon M, Harris KW & Selander KS (2007) Toll like receptor-9 agonists stimulate prostate cancer invasion in vitro. Prostate 67(7): 774–781.

Ilyas M & Tomlinson IP (1997) The interactions of APC, E-cadherin and beta-catenin in tumour development and progression. J Pathol 182(2): 128–137.

Itoh K, Chiba T, Takahashi S, Ishii T, Igarashi K, Katoh Y, Oyake T, Hayashi N, Satoh K, Hatayama I, Yamamoto M & Nabeshima Y (1997) An Nrf2/small Maf heterodimer mediates the induction of phase II detoxifying enzyme genes through antioxidant response elements. Biochem Biophys Res Commun 236(2): 313–322.

Itoh K, Tong KI & Yamamoto M (2004) Molecular mechanism activating Nrf2-Keap1 pathway in regulation of adaptive response to electrophiles. Free Radic Biol Med 36(10): 1208–1213.

Itoh K, Wakabayashi N, Katoh Y, Ishii T, Igarashi K, Engel JD & Yamamoto M (1999) Keap1 represses nuclear activation of antioxidant responsive elements by Nrf2 through binding to the amino-terminal Neh2 domain. Genes Dev 13(1): 76–86.

Iwamoto M, Ahnen DJ, Franklin WA & Maltzman TH (2000) Expression of beta-catenin and full-length APC protein in normal and neoplastic colonic tissues. Carcinogenesis 21(11): 1935–1940.

Iwamura M, Wu W, Muramoto M, Ohori M, Egawa S, Uchida T & Baba S (1999) Parathyroid hormone-related protein is an independent prognostic factor for renal cell carcinoma. Cancer 86(6): 1028–1034.

Jacobsen J, Grankvist K, Rasmuson T, Bergh A, Landberg G & Ljungberg B (2004) Expression of vascular endothelial growth factor protein in human renal cell carcinoma. BJU Int 93(3): 297–302.

Janzen N, Zisman A, Pantuck AJ, Perry K, Schulam P & Belldegrun AS (2002) Minimally invasive ablative approaches in the treatment of renal cell carcinoma. Curr Urol Rep 3(1): 13–20.

117

Jayson M & Sanders H (1998) Increased incidence of serendipitously discovered renal cell carcinoma. Urology 51(2): 203–205.

Jbireal JM, Strell C, Niggemann B, Zanker K & Entschladen F (2010) The selective role of myosin VI in lymphoid leukemia cell migration. Leuk Res 34(12): 1656–1662.

Jin L, Hemperly JJ & Lloyd RV (1991) Expression of neural cell adhesion molecule in normal and neoplastic human neuroendocrine tissues. Am J Pathol 138(4): 961–969.

Johnson DB & Nakada SY (2001) Cryosurgery and needle ablation of renal lesions. J Endourol 15(4): 361–368; discussion 375–376.

Jukkola-Vuorinen A, Rahko E, Vuopala KS, Desmond R, Lehenkari PP, Harris KW & Selander KS (2008) Toll-like receptor-9 expression is inversely correlated with estrogen receptor status in breast cancer. J Innate Immun 1(1): 59–68.

Julius D, Livelli TJ, Jessell TM & Axel R (1989) Ectopic expression of the serotonin 1c receptor and the triggering of malignant transformation. Science 244(4908): 1057–1062.

Jungwirth N, Haeberle L, Schrott KM, Wullich B & Krause FS (2008) Serotonin used as prognostic marker of urological tumors. World J Urol 26(5): 499–504.

Kadmon D, Thompson TC, Lynch GR & Scardino PT (1991) Elevated plasma chromogranin-A concentrations in prostatic carcinoma. J Urol 146(2): 358–361.

Kaelin WG Jr (2002) Molecular basis of the VHL hereditary cancer syndrome. Nat Rev Cancer 2(9): 673–682.

Kaelin WG Jr (2004) The von Hippel-Lindau tumor suppressor gene and kidney cancer. Clin Cancer Res 10(18 Pt 2): 6290S–6295S.

Kallakury BV, Karikehalli S, Haholu A, Sheehan CE, Azumi N & Ross JS (2001) Increased expression of matrix metalloproteinases 2 and 9 and tissue inhibitors of metalloproteinases 1 and 2 correlate with poor prognostic variables in renal cell carcinoma. Clin Cancer Res 7(10): 3113–3119.

Kallio JP, Hirvikoski P, Helin H, Luukkaala T, Tammela TL, Kellokumpu-Lehtinen P & Martikainen PM (2004) Renal cell carcinoma MIB-1, Bax and Bcl-2 expression and prognosis. J Urol 172(6 Pt 1): 2158–2161.

Kallio JP, Tammela TL, Marttinen AT & Kellokumpu-Lehtinen PL (2001) Soluble immunological parameters and early prognosis of renal cell cancer patients. J Exp Clin Cancer Res 20(4): 523–528.

Kankuri-Tammilehto MK, Söderström KO, Pelliniemi TT, Vahlberg T, Pyrhönen SO & Salminen EK (2010) Prognostic evaluation of COX-2 expression in renal cell carcinoma. Anticancer Res 30(7): 3023–3030.

Karakiewicz PI, Briganti A, Chun FK, Trinh QD, Perrotte P, Ficarra V, Cindolo L, De la Taille A, Tostain J, Mulders PF, Salomon L, Zigeuner R, Prayer-Galetti T, Chautard D, Valeri A, Lechevallier E, Descotes JL, Lang H, Méjean A & Patard JJ (2007) Multi-institutional validation of a new renal cancer-specific survival nomogram. J Clin Oncol 25(11): 1316–1322.

Karim R, Tse G, Putti T, Scolyer R & Lee S (2004) The significance of the Wnt pathway in the pathology of human cancers. Pathology 36(2): 120–128.

118

Kaspar JW, Niture SK & Jaiswal AK (2009) Nrf2:INrf2 (Keap1) signaling in oxidative stress. Free Radic Biol Med 47(9): 1304–1309.

Katagiri A, Watanabe R & Tomita Y (1995) E-cadherin expression in renal cell cancer and its significance in metastasis and survival. Br J Cancer 71(2): 376–379.

Kattan MW, Reuter V, Motzer RJ, Katz J & Russo P (2001) A postoperative prognostic nomogram for renal cell carcinoma. J Urol 166(1): 63–67.

Kawai T & Akira S (2008) Toll-like receptor and RIG-I-like receptor signaling. Ann N Y Acad Sci 1143: 1–20.

Kawakami K, Hirata H, Yamamura S, Kikuno N, Saini S, Majid S, Tanaka Y, Kawamoto K, Enokida H, Nakagawa M & Dahiya R (2009) Functional significance of Wnt inhibitory factor-1 gene in kidney cancer. Cancer Res 69(22): 8603–8610.

Kayser K, Schmid W, Ebert W & Wiedenmann B (1988) Expression of neuroendocrine markers (neuronspecific enolase, synaptophysin and bombesin) in carcinoma of the lung. Pathol Res Pract 183(4): 412–417.

Kelleher FC, Fennelly D & Rafferty M (2006) Common critical pathways in embryogenesis and cancer. Acta Oncol 45(4): 375–388.

Kellerman KA & Miller KG (1992) An unconventional myosin heavy chain gene from Drosophila melanogaster. J Cell Biol 119(4): 823–834.

Kim HL, Seligson D, Liu X, Janzen N, Bui MH, Yu H, Shi T, Figlin RA, Horvath S & Belldegrun AS (2004) Using protein expressions to predict survival in clear cell renal carcinoma. Clin Cancer Res 10(16): 5464–5471.

Kim MK & Kim S (2002) Immunohistochemical profile of common epithelial neoplasms arising in the kidney. Appl Immunohistochem Mol Morphol 10(4): 332–338.

Kim YR, Oh JE, Kim MS, Kang MR, Park SW, Han JY, Eom HS, Yoo NJ & Lee SH (2010) Oncogenic NRF2 mutations in squamous cell carcinomas of oesophagus and skin. J Pathol 220(4): 446–451.

Kim YS, Kang YK, Kim JB, Han SA, Kim KI & Paik SR (2000) Beta-Catenin Expression and Mutational Analysis in Renal Cell Carcinomas. Pathol Int 50(9): 725–730.

Klatte T, Rao PN, de Martino M, LaRochelle J, Shuch B, Zomorodian N, Said J, Kabbinavar FF, Belldegrun AS & Pantuck AJ (2009a) Cytogenetic profile predicts prognosis of patients with clear cell renal cell carcinoma. J Clin Oncol 27(5): 746–753.

Klatte T, Seligson DB, LaRochelle J, Shuch B, Said JW, Riggs SB, Zomorodian N, Kabbinavar FF, Pantuck AJ & Belldegrun AS (2009b) Molecular signatures of localized clear cell renal cell carcinoma to predict disease-free survival after nephrectomy. Cancer Epidemiol Biomarkers Prev 18(3): 894–900.

Klatte T, Seligson DB, Leppert JT, Riggs SB, Yu H, Zomorodian N, Kabbinavar FF, Strieter RM, Belldegrun AS & Pantuck AJ (2008) The chemokine receptor CXCR3 is an independent prognostic factor in patients with localized clear cell renal cell carcinoma. J Urol 179(1): 61–66.

Klatte T, Seligson DB, Riggs SB, Leppert JT, Berkman MK, Kleid MD, Yu H, Kabbinavar FF, Pantuck AJ & Belldegrun AS (2007) Hypoxia-inducible factor 1 alpha in clear cell renal cell carcinoma. Clin Cancer Res 13(24): 7388–7393.

119

Klaunig JE, Xu Y, Isenberg JS, Bachowski S, Kolaja KL, Jiang J, Stevenson DE & Walborg EF Jr (1998) The role of oxidative stress in chemical carcinogenesis. Environ Health Perspect 106 Suppl 1: 289–295.

Knudsen B (2006) Migrating with myosin VI. Am J Pathol 169(5): 1523–1526. Kojima T, Shimazui T, Hinotsu S, Joraku A, Oikawa T, Kawai K, Horie R, Suzuki H,

Nagashima R, Yoshikawa K, Michiue T, Asashima M, Akaza H & Uchida K (2009) Decreased expression of CXXC4 promotes a malignant phenotype in renal cell carcinoma by activating Wnt signaling. Oncogene 28(2): 297–305.

Koukoulis GK, Patriarca C & Gould VE (1998) Adhesion molecules and tumor metastasis. Hum Pathol 29(9): 889–892.

Kovács G, Akhtar M, Beckwith BJ, Bugert P, Cooper CS, Delahunt B, Eble JN, Fleming S, Ljungberg B, Medeiros LJ, Moch H, Reuter VE, Ritz E, Roos G, Schmidt D, Srigley JR, Störkel S, van den Berg E & Zbar B (1997) The Heidelberg classification of renal cell tumours. J Pathol 183(2): 131–133.

Krieg AM (2008) Toll-like receptor 9 (TLR9) agonists in the treatment of cancer. Oncogene 27(2): 161–167.

Krishnamachary B, Zagzag D, Nagasawa H, Rainey K, Okuyama H, Baek JH & Semenza GL (2006) Hypoxia-inducible factor-1-dependent repression of E-cadherin in von Hippel-Lindau tumor suppressor-null renal cell carcinoma mediated by TCF3, ZFHX1A, and ZFHX1B. Cancer Res 66(5): 2725–2731.

Kuhlicke J, Frick JS, Morote-Garcia JC, Rosenberger P & Eltzschig HK (2007) Hypoxia inducible factor (HIF)-1 coordinates induction of Toll-like receptors TLR2 and TLR6 during hypoxia. PLoS One 2(12): e1364.

Kumar B, Koul S, Khandrika L, Meacham RB & Koul HK (2008) Oxidative stress is inherent in prostate cancer cells and is required for aggressive phenotype. Cancer Res 68(6): 1777–1785.

Kuroda N, Hes O, Miyazaki E, Shuin T & Enzan H (2006) Frequent expression of neuroendocrine markers in mucinous tubular and spindle cell carcinoma of the kidney. Histol Histopathol 21(1): 7–10.

Kuroiwa K, Konomoto T, Kumazawa J, Naito S & Tsuneyoshi M (2001) Cell proliferative activity and expression of cell-cell adhesion factors (E-cadherin, alpha-, beta-, and gamma-catenin, and p120) in sarcomatoid renal cell carcinoma. J Surg Oncol 77(2): 123–131.

Küssel-Andermann P, El-Amraoui A, Safieddine S, Nouaille S, Perfettini I, Lecuit M, Cossart P, Wolfrum U & Petit C (2000) Vezatin, a novel transmembrane protein, bridges myosin VIIA to the cadherin-catenins complex. EMBO J 19(22): 6020–6029.

Lahr G, Mayerhofer A, Bucher S, Barthels D, Wille W & Gratzl M (1993) Neural cell adhesion molecules in rat endocrine tissues and tumor cells: distribution and molecular analysis. Endocrinology 132(3): 1207–1217.

Lam JS, Klatte T, Kim HL, Patard JJ, Breda A, Zisman A, Pantuck AJ & Figlin RA (2008) Prognostic factors and selection for clinical studies of patients with kidney cancer. Crit Rev Oncol Hematol 65(3): 235–262.

120

Lamphier MS, Sirois CM, Verma A, Golenbock DT & Latz E (2006) TLR9 and the recognition of self and non-self nucleic acids. Ann N Y Acad Sci 1082: 31–43.

Landis SH, Murray T, Bolden S & Wingo PA (1999) Cancer statistics, 1999. CA Cancer J Clin 49(1): 8–31.

Lane BR, Chery F, Jour G, Sercia L, Magi-Galluzzi C, Novick AC & Zhou M (2007) Renal neuroendocrine tumours: a clinicopathological study. BJU Int 100(5): 1030–1035.

Lang H, Lindner V, Letourneux H, Martin M, Saussine C & Jacqmin D (2004) Prognostic value of microscopic venous invasion in renal cell carcinoma: long-term follow-up. Eur Urol 46(3): 331–335.

Latif F, Tory K, Gnarra J, Yao M, Duh FM, Orcutt ML, Stackhouse T, Kuzmin I, Modi W & Geil L (1993) Identification of the von Hippel-Lindau disease tumor suppressor gene. Science 260(5112): 1317–1320.

Laurent TC, Moore EC & Reichard P (1964) Enzymatic Synthesis of Deoxyribonucleotides. Iv. Isolation and Characterization of Thioredoxin, the Hydrogen Donor from Escherichia Coli B. J Biol Chem 239: 3436–3444.

Leifer CA, Kennedy MN, Mazzoni A, Lee C, Kruhlak MJ & Segal DM (2004) TLR9 is localized in the endoplasmic reticulum prior to stimulation. J Immunol 173(2): 1179–1183.

Levy NS, Chung S, Furneaux H & Levy AP (1998) Hypoxic stabilization of vascular endothelial growth factor mRNA by the RNA-binding protein HuR. J Biol Chem 273(11): 6417–6423.

Li FP, Decker HJ, Zbar B, Stanton VP Jr, Kovács G, Seizinger BR, Aburatani H, Sandberg AA, Berg S, Hosoe S & Brown RS (1993) Clinical and genetic studies of renal cell carcinomas in a family with a constitutional chromosome 3;8 translocation. Genetics of familial renal carcinoma. Ann Intern Med 118(2): 106–111.

Lidgren A, Hedberg Y, Grankvist K, Rasmuson T, Vasko J & Ljungberg B (2005) The expression of hypoxia-inducible factor 1alpha is a favorable independent prognostic factor in renal cell carcinoma. Clin Cancer Res 11(3): 1129–1135.

Lim SJ, Kim HJ, Kim JY, Park K & Lee CM (2007) Expression of HuR is associated with increased cyclooxygenase-2 expression in uterine cervical carcinoma. Int J Gynecol Pathol 26(3): 229–234.

Linden MD, Torres FX, Kubus J & Zarbo RJ (1992) Clinical application of morphologic and immunocytochemical assessments of cell proliferation. Am J Clin Pathol 97(5 Suppl 1): S4–13.

Linehan WM, Walther MM & Zbar B (2003) The genetic basis of cancer of the kidney. J Urol 170(6 Pt 1): 2163–2172.

Link RE, Bhayani SB, Allaf ME, Varkarakis I, Inagaki T, Rogers C, Su LM, Jarrett TW & Kavoussi LR (2005) Exploring the learning curve, pathological outcomes and perioperative morbidity of laparoscopic partial nephrectomy performed for renal mass. J Urol 173(5): 1690–1694.

Liou GY & Storz P (2010) Reactive oxygen species in cancer. Free Radic Res 44(5): 479–496.

121

Lipworth L, Tarone RE & McLaughlin JK (2006) The epidemiology of renal cell carcinoma. J Urol 176(6 Pt 1): 2353–2358.

Liu CH, Chang SH, Narko K, Trifan OC, Wu MT, Smith E, Haudenschild C, Lane TF & Hla T (2001) Overexpression of cyclooxygenase-2 is sufficient to induce tumorigenesis in transgenic mice. J Biol Chem 276(21): 18563–18569.

Liu H, Colavitti R, Rovira II & Finkel T (2005) Redox-dependent transcriptional regulation. Circ Res 97(10): 967–974.

Liu Y, Zhu L, Fatheree NY, Liu X, Pacheco SE, Tatevian N & Rhoads JM (2009) Changes in intestinal Toll-like receptors and cytokines precede histological injury in a rat model of necrotizing enterocolitis. Am J Physiol Gastrointest Liver Physiol 297(3): G442–450.

Ljungberg B, Alamdari FI, Stenling R & Roos G (1999) Prognostic significance of the Heidelberg classification of renal cell carcinoma. Eur Urol 36(6): 565–569.

Ljungberg B, Cowan NC, Hanbury DC, Hora M, Kuczyk MA, Merseburger AS, Patard JJ, Mulders PF, Sinescu IC & European Association of Urology Guideline Group (2010) EAU guidelines on renal cell carcinoma: the 2010 update. Eur Urol 58(3): 398–406.

Loikkanen I, Toljamo K, Hirvikoski P, Väisänen T, Paavonen TK & Vaarala MH (2009) Myosin VI is a modulator of androgen-dependent gene expression. Oncol Rep 22(5): 991–995.

Lokich J (1997) Spontaneous regression of metastatic renal cancer. Case report and literature review. Am J Clin Oncol 20(4): 416–418.

López de Silanes I, Fan J, Yang X, Zonderman AB, Potapova O, Pizer ES & Gorospe M (2003) Role of the RNA-binding protein HuR in colon carcinogenesis. Oncogene 22(46): 7146–7154.

López de Silanes I, Lal A & Gorospe M (2005) HuR: post-transcriptional paths to malignancy. RNA Biol 2(1): 11–13.

Lui KW, Gervais DA, Arellano RA & Mueller PR (2003) Radiofrequency ablation of renal cell carcinoma. Clin Radiol 58(12): 905–913.

Ma WJ, Cheng S, Campbell C, Wright A & Furneaux H (1996) Cloning and characterization of HuR, a ubiquitously expressed Elav-like protein. J Biol Chem 271(14): 8144–8151.

Maher ER, Iselius L, Yates JR, Littler M, Benjamin C, Harris R, Sampson J, Williams A, Ferguson-Smith MA & Morton N (1991) Von Hippel-Lindau disease: a genetic study. J Med Genet 28(7): 443–447.

Makretsov N, Gilks CB, Coldman AJ, Hayes M & Huntsman D (2003) Tissue microarray analysis of neuroendocrine differentiation and its prognostic significance in breast cancer. Hum Pathol 34(10): 1001–1008.

Makrilia N, Kollias A, Manolopoulos L & Syrigos K (2009) Cell adhesion molecules: role and clinical significance in cancer. Cancer Invest 27(10): 1023–1037.

Marshak-Rothstein A & Rifkin IR (2007) Immunologically active autoantigens: the role of toll-like receptors in the development of chronic inflammatory disease. Annu Rev Immunol 25: 419–441.

122

Masferrer JL, Leahy KM, Koki AT, Zweifel BS, Settle SL, Woerner BM, Edwards DA, Flickinger AG, Moore RJ & Seibert K (2000) Antiangiogenic and antitumor activities of cyclooxygenase-2 inhibitors. Cancer Res 60(5): 1306–1311.

Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, Wykoff CC, Pugh CW, Maher ER & Ratcliffe PJ (1999) The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399(6733): 271–275.

McMahon M, Itoh K, Yamamoto M & Hayes JD (2003) Keap1-dependent proteasomal degradation of transcription factor Nrf2 contributes to the negative regulation of antioxidant response element-driven gene expression. J Biol Chem 278(24): 21592–21600.

Méjean A, Oudard S & Thiounn N (2003) Prognostic factors of renal cell carcinoma. J Urol 169(3): 821–827.

Mempel M, Voelcker V, Köllisch G, Plank C, Rad R, Gerhard M, Schnopp C, Fraunberger P, Walli AK, Ring J, Abeck D & Ollert M (2003) Toll-like receptor expression in human keratinocytes: nuclear factor kappaB controlled gene activation by Staphylococcus aureus is toll-like receptor 2 but not toll-like receptor 4 or platelet activating factor receptor dependent. J Invest Dermatol 121(6): 1389–1396.

Menter DG, Schilsky RL & DuBois RN (2010) Cyclooxygenase-2 and cancer treatment: understanding the risk should be worth the reward. Clin Cancer Res 16(5): 1384–1390.

Merikallio H, Pääkkö P, Kinnula VL, Harju T & Soini Y (2011) Nuclear factor erythroid-derived 2-like 2 (Nrf2) and DJ1 are prognostic factors in lung cancer. Hum Pathol Epub Sep 22.

Merrell MA, Ilvesaro JM, Lehtonen N, Sorsa T, Gehrs B, Rosenthal E, Chen D, Shackley B, Harris KW & Selander KS (2006) Toll-like receptor 9 agonists promote cellular invasion by increasing matrix metalloproteinase activity. Mol Cancer Res 4(7): 437–447.

Merseburger AS, Anastasiadis AG, Hennenlotter J, Schilling D, Simon P, Machtens SA, Serth J, Stenzl A & Kuczyk MA (2006) Tissue microarrays: applications in urological cancer research. World J Urol 24(5): 579–584.

Mevorach RA, Segal AJ, Tersegno ME & Frank IN (1992) Renal cell carcinoma: incidental diagnosis and natural history: review of 235 cases. Urology 39(6): 519–522.

Meyer T & Hart IR (1998) Mechanisms of tumour metastasis. Eur J Cancer 34(2): 214–221.

Migita T, Oda Y, Naito S & Tsuneyoshi M (2002) Low expression of p27(Kip1) is associated with tumor size and poor prognosis in patients with renal cell carcinoma. Cancer 94(4): 973–979.

Millo H, Leaper K, Lazou V & Bownes M (2004) Myosin VI plays a role in cell-cell adhesion during epithelial morphogenesis. Mech Dev 121(11): 1335–1351.

Miyata Y, Koga S, Kanda S, Nishikido M, Hayashi T & Kanetake H (2003) Expression of cyclooxygenase-2 in renal cell carcinoma: correlation with tumor cell proliferation, apoptosis, angiogenesis, expression of matrix metalloproteinase-2, and survival. Clin Cancer Res 9(5): 1741–1749.

123

Mizutani Y, Nakanishi H, Yamamoto K, Li YN, Matsubara H, Mikami K, Okihara K, Kawauchi A, Bonavida B & Miki T (2005) Downregulation of Smac/DIABLO expression in renal cell carcinoma and its prognostic significance. J Clin Oncol 23(3): 448–454.

Moch H, Artibani W, Delahunt B, Ficarra V, Knuechel R, Montorsi F, Patard JJ, Stief CG, Sulser T & Wild PJ (2009) Reassessing the current UICC/AJCC TNM staging for renal cell carcinoma. Eur Urol 56(4): 636–643.

Montero-Hadjadje M, Vaingankar S, Elias S, Tostivint H, Mahata SK & Anouar Y (2008) Chromogranins A and B and secretogranin II: evolutionary and functional aspects. Acta Physiol (Oxf) 192(2): 309–324.

Motohashi H & Yamamoto M (2004) Nrf2-Keap1 defines a physiologically important stress response mechanism. Trends Mol Med 10(11): 549–557.

Motzer RJ, Bander NH & Nanus DM (1996) Renal-cell carcinoma. N Engl J Med 335(12): 865–875.

Motzer RJ, Mazumdar M, Bacik J, Berg W, Amsterdam A & Ferrara J (1999) Survival and prognostic stratification of 670 patients with advanced renal cell carcinoma. J Clin Oncol 17(8): 2530–2540.

Mrena J, Wiksten JP, Thiel A, Kokkola A, Pohjola L, Lundin J, Nordling S, Ristimäki A & Haglund C (2005) Cyclooxygenase-2 is an independent prognostic factor in gastric cancer and its expression is regulated by the messenger RNA stability factor HuR. Clin Cancer Res 11(20): 7362–7368.

Mungan MU, Gurel D, Canda AE, Tuna B, Yorukoglu K & Kirkali Z (2006) Expression of COX-2 in normal and pyelonephritic kidney, renal intraepithelial neoplasia, and renal cell carcinoma. Eur Urol 50(1): 92–7; discussion 97.

Murphy AM, Gilbert SM, Katz AE, Goluboff ET, Sawczuk IS, Olsson CA, Benson MC & McKiernan JM (2005) Re-evaluation of the Tumour-Node-Metastasis staging of locally advanced renal cortical tumours: absolute size (T2) is more significant than renal capsular invasion (T3a). BJU Int 95(1): 27–30.

Nabors LB, Gillespie GY, Harkins L & King PH (2001) HuR, a RNA stability factor, is expressed in malignant brain tumors and binds to adenine- and uridine-rich elements within the 3' untranslated regions of cytokine and angiogenic factor mRNAs. Cancer Res 61(5): 2154–2161.

Namdarian B, Tan KV, Fankhauser MJ, Nguyen TT, Corcoran NM, Costello AJ & Hovens CM (2010) Circulating endothelial cells and progenitors: potential biomarkers of renal cell carcinoma. BJU Int 106(7): 1081–1087.

Nesland JM, Holm R, Johannessen JV & Gould VE (1988) Neuroendocrine differentiation in breast lesions. Pathol Res Pract 183(2): 214–221.

Neumann E, Engelsberg A, Decker J, Störkel S, Jaeger E, Huber C & Seliger B (1998) Heterogeneous expression of the tumor-associated antigens RAGE-1, PRAME, and glycoprotein 75 in human renal cell carcinoma: candidates for T-cell-based immunotherapies? Cancer Res 58(18): 4090–4095.

Nioi P & Nguyen T (2007) A mutation of Keap1 found in breast cancer impairs its ability to repress Nrf2 activity. Biochem Biophys Res Commun 362(4): 816–821.

124

Nishiya T & DeFranco AL (2004) Ligand-regulated chimeric receptor approach reveals distinctive subcellular localization and signaling properties of the Toll-like receptors. J Biol Chem 279(18): 19008–19017.

Nisman B, Yutkin V, Nechushtan H, Gofrit ON, Peretz T, Gronowitz S & Pode D (2010) Circulating tumor M2 pyruvate kinase and thymidine kinase 1 are potential predictors for disease recurrence in renal cell carcinoma after nephrectomy. Urology 76(2): 513.e1–513.e6.

Niture SK & Jaiswal AK (2011) INrf2 (Keap1) targets Bcl-2 degradation and controls cellular apoptosis. Cell Death Differ 18(3): 439–451.

Nogueira M & Kim HL (2008) Molecular markers for predicting prognosis of renal cell carcinoma. Urol Oncol 26(2): 113–124.

Oberley TD (2002) Oxidative damage and cancer. Am J Pathol 160(2): 403–408. Ohta T, Iijima K, Miyamoto M, Nakahara I, Tanaka H, Ohtsuji M, Suzuki T, Kobayashi A,

Yokota J, Sakiyama T, Shibata T, Yamamoto M & Hirohashi S (2008) Loss of Keap1 function activates Nrf2 and provides advantages for lung cancer cell growth. Cancer Res 68(5): 1303–1309.

Okamoto K, Toyokuni S, Uchida K, Ogawa O, Takenewa J, Kakehi Y, Kinoshita H, Hattori-Nakakuki Y, Hiai H & Yoshida O (1994) Formation of 8-hydroxy-2'-deoxyguanosine and 4-hydroxy-2-nonenal-modified proteins in human renal-cell carcinoma. Int J Cancer 58(6): 825–829.

Orsulic S, Huber O, Aberle H, Arnold S & Kemler R (1999) E-cadherin binding prevents beta-catenin nuclear localization and beta-catenin/LEF-1-mediated transactivation. J Cell Sci 112 (Pt 8)(Pt 8): 1237–1245.

Ougolkov A, Mai M, Takahashi Y, Omote K, Bilim V, Shimizu A & Minamoto T (2000) Altered expression of beta-catenin and c-erbB-2 in early gastric cancer. J Exp Clin Cancer Res 19(3): 349–355.

Pagani A, Papotti M, Höfler H, Weiler R, Winkler H & Bussolati G (1990) Chromogranin A and B gene expression in carcinomas of the breast. Correlation of immunocytochemical, immunoblot, and hybridization analyses. Am J Pathol 136(2): 319–327.

Pantuck AJ, Seligson DB, Klatte T, Yu H, Leppert JT, Moore L, O'Toole T, Gibbons J, Belldegrun AS & Figlin RA (2007) Prognostic relevance of the mTOR pathway in renal cell carcinoma: implications for molecular patient selection for targeted therapy. Cancer 109(11): 2257–2267.

Pantuck AJ, Zisman A & Belldegrun AS (2001) The changing natural history of renal cell carcinoma. J Urol 166(5): 1611–1623.

Pantuck AJ, Zisman A, Rauch MK & Belldegrun A (2000) Incidental renal tumors. Urology 56(2): 190–196.

Papadimitraki ED, Tzardi M, Bertsias G, Sotsiou E & Boumpas DT (2009) Glomerular expression of toll-like receptor-9 in lupus nephritis but not in normal kidneys: implications for the amplification of the inflammatory response. Lupus 18(9): 831–835.

125

Park JG, Choe GY, Helman LJ, Gazdar AF, Yang HK, Kim JP, Park SH & Kim YI (1992) Chromogranin-A expression in gastric and colon cancer tissues. Int J Cancer 51(2): 189–194.

Parker AS, Cheville JC, Janney CA & Cerhan JR (2002) High expression levels of insulin-like growth factor-I receptor predict poor survival among women with clear-cell renal cell carcinomas. Hum Pathol 33(8): 801–805.

Parker AS, Kosari F, Lohse CM, Houston Thompson R, Kwon ED, Murphy L, Riehle DL, Blute ML, Leibovich BC, Vasmatzis G & Cheville JC (2006) High expression levels of survivin protein independently predict a poor outcome for patients who undergo surgery for clear cell renal cell carcinoma. Cancer 107(1): 37–45.

Parker AS, Leibovich BC, Lohse CM, Sheinin Y, Kuntz SM, Eckel-Passow JE, Blute ML & Kwon ED (2009) Development and evaluation of BioScore: a biomarker panel to enhance prognostic algorithms for clear cell renal cell carcinoma. Cancer 115(10): 2092–2103.

Parkin DM, Bray F, Ferlay J & Pisani P (2005) Global cancer statistics, 2002. CA Cancer J Clin 55(2): 74–108.

Patard JJ, Shvarts O, Lam JS, Pantuck AJ, Kim HL, Ficarra V, Cindolo L, Han KR, De La Taille A, Tostain J, Artibani W, Abbou CC, Lobel B, Chopin DK, Figlin RA, Mulders PF & Belldegrun AS (2004) Safety and efficacy of partial nephrectomy for all T1 tumors based on an international multicenter experience. J Urol 171(6 Pt 1): 2181–2185.

Peng SS, Chen CY, Xu N & Shyu AB (1998) RNA stabilization by the AU-rich element binding protein, HuR, an ELAV protein. EMBO J 17(12): 3461–3470.

Peruzzi B, Athauda G & Bottaro DP (2006) The von Hippel-Lindau tumor suppressor gene product represses oncogenic beta-catenin signaling in renal carcinoma cells. Proc Natl Acad Sci U S A 103(39): 14531–14536.

Pfaffenroth EC & Linehan WM (2008) Genetic basis for kidney cancer: opportunity for disease-specific approaches to therapy. Expert Opin Biol Ther 8(6): 779–790.

Platz J, Beisswenger C, Dalpke A, Koczulla R, Pinkenburg O, Vogelmeier C & Bals R (2004) Microbial DNA induces a host defense reaction of human respiratory epithelial cells. J Immunol 173(2): 1219–1223.

Polcari AJ, Gorbonos A, Milner JE & Flanigan RC (2009) The role of cytoreductive nephrectomy in the era of molecular targeted therapy. Int J Urol 16(3): 227–233.

Powis G & Montfort WR (2001) Properties and biological activities of thioredoxins. Annu Rev Pharmacol Toxicol 41: 261–295.

Powis G, Mustacich D & Coon A (2000) The role of the redox protein thioredoxin in cell growth and cancer. Free Radic Biol Med 29(3–4): 312–322.

Prescott SM & White RL (1996) Self-promotion? Intimate connections between APC and prostaglandin H synthase-2. Cell 87(5): 783–786.

Puri C, Chibalina MV, Arden SD, Kruppa AJ, Kendrick-Jones J & Buss F (2010) Overexpression of myosin VI in prostate cancer cells enhances PSA and VEGF secretion, but has no effect on endocytosis. Oncogene 29(2): 188–200.

126

Raffel J, Bhattacharyya AK, Gallegos A, Cui H, Einspahr JG, Alberts DS & Powis G (2003) Increased expression of thioredoxin-1 in human colorectal cancer is associated with decreased patient survival. J Lab Clin Med 142(1): 46–51.

Rasmuson T, Grankvist K & Ljungberg B (1993) Serum gamma-enolase and prognosis of patients with renal cell carcinoma. Cancer 72(4): 1324–1328.

Rasmuson T, Grankvist K, Roos G & Ljungberg B (1999) Neuroendocrine differentiation in renal cell carcinoma-evaluation of chromogranin A and neuron-specific enolase. Acta Oncol 38(5): 623–628.

Rayburn ER, Wang W, Zhang Z, Li M, Zhang R & Wang H (2006) Experimental therapy of prostate cancer with an immunomodulatory oligonucleotide: effects on tumor growth, apoptosis, proliferation, and potentiation of chemotherapy. Prostate 66(15): 1653–1663.

Rayman P, Wesa AK, Richmond AL, Das T, Biswas K, Raval G, Storkus WJ, Tannenbaum C, Novick A, Bukowski R & Finke J (2004) Effect of renal cell carcinomas on the development of type 1 T-cell responses. Clin Cancer Res 10(18 Pt 2): 6360S–6366S.

Reddy SR & Reddy RK (2010) Immunology and Immune System. In: A textbook of microbiology, volume III: Immunology & medical microbiology. Mumbai, Himalaya Publishing House: 6–22.

Ren T, Wen ZK, Liu ZM, Liang YJ, Guo ZL & Xu L (2007) Functional expression of TLR9 is associated to the metastatic potential of human lung cancer cell: functional active role of TLR9 on tumor metastasis. Cancer Biol Ther 6(11): 1704–1709.

Rini BI, Weinberg V, Dunlap S, Elchinoff A, Yu N, Bok R, Simko J & Small EJ (2006) Maximal COX-2 immunostaining and clinical response to celecoxib and interferon alpha therapy in metastatic renal cell carcinoma. Cancer 106(3): 566–575.

Ristimäki A, Honkanen N, Jänkälä H, Sipponen P & Härkönen M (1997) Expression of cyclooxygenase-2 in human gastric carcinoma. Cancer Res 57(7): 1276–1280.

Ristimäki A, Sivula A, Lundin J, Lundin M, Salminen T, Haglund C, Joensuu H & Isola J (2002) Prognostic significance of elevated cyclooxygenase-2 expression in breast cancer. Cancer Res 62(3): 632–635.

Roberts R, Lister I, Schmitz S, Walker M, Veigel C, Trinick J, Buss F & Kendrick-Jones J (2004) Myosin VI: cellular functions and motor properties. Philos Trans R Soc Lond B Biol Sci 359(1452): 1931–1944.

Robson CJ, Churchill BM & Anderson W (1969) The results of radical nephrectomy for renal cell carcinoma. J Urol 101(3): 297–301.

Ruan K, Song G & Ouyang G (2009) Role of hypoxia in the hallmarks of human cancer. J Cell Biochem 107(6): 1053–1062.

Rubartelli A, Bajetto A, Allavena G, Wollman E & Sitia R (1992) Secretion of thioredoxin by normal and neoplastic cells through a leaderless secretory pathway. J Biol Chem 267(34): 24161–24164.

Rubartelli A, Bonifaci N & Sitia R (1995) High rates of thioredoxin secretion correlate with growth arrest in hepatoma cells. Cancer Res 55(3): 675–680.

127

Russell RC & Ohh M (2007) The role of VHL in the regulation of E-cadherin: a new connection in an old pathway. Cell Cycle 6(1): 56–59.

Saegusa M, Hashimura M, Yoshida T & Okayasu I (2001) Beta-catenin mutations and aberrant nuclear expression during endometrial tumorigenesis. Br J Cancer 84(2): 209–217.

Satoh H, Moriguchi T, Taguchi K, Takai J, Maher JM, Suzuki T, Winnard PT,Jr, Raman V, Ebina M, Nukiwa T & Yamamoto M (2010) Nrf2-deficiency creates a responsive microenvironment for metastasis to the lung. Carcinogenesis 31(10): 1833–1843.

Schaefer TM, Desouza K, Fahey JV, Beagley KW & Wira CR (2004) Toll-like receptor (TLR) expression and TLR-mediated cytokine/chemokine production by human uterine epithelial cells. Immunology 112(3): 428–436.

Schips L, Lipsky K, Zigeuner R, Salfellner M, Winkler S, Langner C, Rehak P, Pummer K & Hubmer G (2003) Impact of tumor-associated symptoms on the prognosis of patients with renal cell carcinoma: a single-center experience of 683 patients. Urology 62(6): 1024–1028.

Schmausser B, Andrulis M, Endrich S, Lee SK, Josenhans C, Müller-Hermelink HK & Eck M (2004) Expression and subcellular distribution of toll-like receptors TLR4, TLR5 and TLR9 on the gastric epithelium in Helicobacter pylori infection. Clin Exp Immunol 136(3): 521–526.

Schmausser B, Andrulis M, Endrich S, Müller-Hermelink HK & Eck M (2005) Toll-like receptors TLR4, TLR5 and TLR9 on gastric carcinoma cells: an implication for interaction with Helicobacter pylori. Int J Med Microbiol 295(3): 179–185.

Schmechel D, Marangos PJ, Zis AP, Brightman M & Goodwin FK (1978) Brain endolases as specific markers of neuronal and glial cells. Science 199(4326): 313–315.

Schmidt L, Duh FM, Chen F, Kishida T, Glenn G, Choyke P, Scherer SW, Zhuang Z, Lubensky I, Dean M, Allikmets R, Chidambaram A, Bergerheim UR, Feltis JT, Casadevall C, Zamarron A, Bernues M, Richard S, Lips CJ, Walther MM, Tsui LC, Geil L, Orcutt ML, Stackhouse T, Lipan J, Slife L, Brauch H, Decker J, Niehans G, Hughson MD, Moch H, Störkel S, Lerman MI, Linehan WM & Zbar B (1997) Germline and somatic mutations in the tyrosine kinase domain of the MET proto-oncogene in papillary renal carcinomas. Nat Genet 16(1): 68–73.

Sengupta S, Jang BC, Wu MT, Paik JH, Furneaux H & Hla T (2003) The RNA-binding protein HuR regulates the expression of cyclooxygenase-2. J Biol Chem 278(27): 25227–25233.

Serra S, Salahshor S, Fagih M, Niakosari F, Radhi JM & Chetty R (2007) Nuclear expression of E-cadherin in solid pseudopapillary tumors of the pancreas. JOP 8(3): 296–303.

Shaw RJ & Cantley LC (2006) Ras, PI(3)K and mTOR signalling controls tumour cell growth. Nature 441(7092): 424–430.

Sheehan KM, Sheahan K, O'Donoghue DP, MacSweeney F, Conroy RM, Fitzgerald DJ & Murray FE (1999) The relationship between cyclooxygenase-2 expression and colorectal cancer. JAMA 282(13): 1254–1257.

128

Sheflin LG, Zou AP & Spaulding SW (2004) Androgens regulate the binding of endogenous HuR to the AU-rich 3'UTRs of HIF-1alpha and EGF mRNA. Biochem Biophys Res Commun 322(2): 644–651.

Sheng H, Williams CS, Shao J, Liang P, DuBois RN & Beauchamp RD (1998) Induction of cyclooxygenase-2 by activated Ha-ras oncogene in Rat-1 fibroblasts and the role of mitogen-activated protein kinase pathway. J Biol Chem 273(34): 22120–22127.

Shi Z, Cai Z, Sanchez A, Zhang T, Wen S, Wang J, Yang J, Fu S & Zhang D (2011) A novel Toll-like receptor that recognizes vesicular stomatitis virus. J Biol Chem 286(6): 4517–4524.

Shibata T, Kokubu A, Gotoh M, Ojima H, Ohta T, Yamamoto M & Hirohashi S (2008a) Genetic alteration of Keap1 confers constitutive Nrf2 activation and resistance to chemotherapy in gallbladder cancer. Gastroenterology 135(4): 1358–1368, 1368.e1–4.

Shibata T, Ohta T, Tong KI, Kokubu A, Odogawa R, Tsuta K, Asamura H, Yamamoto M & Hirohashi S (2008b) Cancer related mutations in NRF2 impair its recognition by Keap1-Cul3 E3 ligase and promote malignancy. Proc Natl Acad Sci U S A 105(36): 13568–13573.

Shibata T, Saito S, Kokubu A, Suzuki T, Yamamoto M & Hirohashi S (2010) Global downstream pathway analysis reveals a dependence of oncogenic NF-E2-related factor 2 mutation on the mTOR growth signaling pathway. Cancer Res 70(22): 9095–9105.

Shiina H, Igawa M, Breault J, Ribeiro-Filho L, Pookot D, Urakami S, Terashima M, Deguchi M, Yamanaka M, Shirai M, Kaneuchi M, Kane CJ & Dahiya R (2003) The human T-cell factor-4 gene splicing isoforms, Wnt signal pathway, and apoptosis in renal cell carcinoma. Clin Cancer Res 9(6): 2121–2132.

Shimazui T, Bringuier PP, van Berkel H, Ruijter E, Akaza H, Debruyne FM, Oosterwijk E & Schalken JA (1997) Decreased expression of alpha-catenin is associated with poor prognosis of patients with localized renal cell carcinoma. Int J Cancer 74(5): 523–528.

Shimazui T, Kojima T, Onozawa M, Suzuki M, Asano T & Akaza H (2006) Expression profile of N-cadherin differs from other classical cadherins as a prognostic marker in renal cell carcinoma. Oncol Rep 15(5): 1181–1184.

Shimazui T, Schalken JA, Giroldi LA, Jansen CF, Akaza H, Koiso K, Debruyne FM & Bringuier PP (1996) Prognostic value of cadherin-associated molecules (alpha-, beta-, and gamma-catenins and p120cas) in bladder tumors. Cancer Res 56(18): 4154–4158.

Shvarts O, Seligson D, Lam J, Shi T, Horvath S, Figlin R, Belldegrun A & Pantuck AJ (2005) P53 is an Independent Predictor of Tumor Recurrence and Progression After Nephrectomy in Patients with Localized Renal Cell Carcinoma. J Urol 173(3): 725–728.

Siddiqui EJ, Thompson CS, Mikhailidis DP & Mumtaz FH (2005) The role of serotonin in tumour growth (review). Oncol Rep 14(6): 1593–1597.

129

Singh A, Boldin-Adamsky S, Thimmulappa RK, Rath SK, Ashush H, Coulter J, Blackford A, Goodman SN, Bunz F, Watson WH, Gabrielson E, Feinstein E & Biswal S (2008) RNAi-mediated silencing of nuclear factor erythroid-2-related factor 2 gene expression in non-small cell lung cancer inhibits tumor growth and increases efficacy of chemotherapy. Cancer Res 68(19): 7975–7984.

Singh A, Misra V, Thimmulappa RK, Lee H, Ames S, Hoque MO, Herman JG, Baylin SB, Sidransky D, Gabrielson E, Brock MV & Biswal S (2006) Dysfunctional KEAP1-NRF2 interaction in non-small-cell lung cancer. PLoS Med 3(10): e420.

Skov BG, Sørensen JB, Hirsch FR, Larsson LI & Hansen HH (1991) Prognostic impact of histologic demonstration of chromogranin A and neuron specific enolase in pulmonary adenocarcinoma. Ann Oncol 2(5): 355–360.

Smith WL, Garavito RM & DeWitt DL (1996) Prostaglandin endoperoxide H synthases (cyclooxygenases)-1 and -2. J Biol Chem 271(52): 33157–33160.

Sobin LH, Gospodarowicz MK & Wittekind C (2009) TNM classification of malignant tumours. Chichester, Wiley-Blackwell.

Sobin LH & Wittekind C (2002) TNM classification of malignant tumours. New York, Wiley-Liss.

Soini Y, Kallio JP, Hirvikoski P, Helin H, Kellokumpu-Lehtinen P, Kang SW, Tammela TL, Peltoniemi M, Martikainen PM & Kinnula VL (2006a) Oxidative/nitrosative stress and peroxiredoxin 2 are associated with grade and prognosis of human renal carcinoma. APMIS 114(5): 329–337.

Soini Y, Kallio JP, Hirvikoski P, Helin H, Kellokumpu-Lehtinen P, Tammela TL, Peltoniemi M, Martikainen PM & Kinnula LV (2006b) Antioxidant enzymes in renal cell carcinoma. Histol Histopathol 21(2): 157–165.

Solis LM, Behrens C, Dong W, Suraokar M, Ozburn NC, Moran CA, Corvalan AH, Biswal S, Swisher SG, Bekele BN, Minna JD, Stewart DJ & Wistuba II (2010) Nrf2 and Keap1 abnormalities in non-small cell lung carcinoma and association with clinicopathologic features. Clin Cancer Res 16(14): 3743–3753.

Sommers CL, Thompson EW, Torri JA, Kemler R, Gelmann EP & Byers SW (1991) Cell adhesion molecule uvomorulin expression in human breast cancer cell lines: relationship to morphology and invasive capacities. Cell Growth Differ 2(8): 365–372.

Spector A, Yan GZ, Huang RR, McDermott MJ, Gascoyne PR & Pigiet V (1988) The effect of H2O2 upon thioredoxin-enriched lens epithelial cells. J Biol Chem 263(10): 4984–4990.

Stacy DR, Ely K, Massion PP, Yarbrough WG, Hallahan DE, Sekhar KR & Freeman ML (2006) Increased expression of nuclear factor E2 p45-related factor 2 (NRF2) in head and neck squamous cell carcinomas. Head Neck 28(9): 813–818.

Subbaramaiah K, Telang N, Ramonetti JT, Araki R, DeVito B, Weksler BB & Dannenberg AJ (1996) Transcription of cyclooxygenase-2 is enhanced in transformed mammary epithelial cells. Cancer Res 56(19): 4424–4429.

Sweeney HL & Houdusse A (2007) What can myosin VI do in cells? Curr Opin Cell Biol 19(1): 57–66.

130

Syversen U, Halvorsen T, Mårvik R & Waldum HL (1995) Neuroendocrine differentiation in colorectal carcinomas. Eur J Gastroenterol Hepatol 7(7): 667–674.

Szatrowski TP & Nathan CF (1991) Production of large amounts of hydrogen peroxide by human tumor cells. Cancer Res 51(3): 794–798.

Taguchi K, Motohashi H & Yamamoto M (2011) Molecular mechanisms of the Keap1-Nrf2 pathway in stress response and cancer evolution. Genes Cells 16(2): 123–140.

Takahashi T, Sonobe M, Menju T, Nakayama E, Mino N, Iwakiri S, Nagai S, Sato K, Miyahara R, Okubo K, Hirata T, Date H & Wada H (2010) Mutations in Keap1 are a potential prognostic factor in resected non-small cell lung cancer. J Surg Oncol 101(6): 500–506.

Takashi M, Haimoto H, Tanaka J, Murase T & Kato K (1989) Evaluation of gamma-enolase as a tumor marker for renal cell carcinoma. J Urol 141(4): 830–834.

Takayama T, Shiozaki H, Shibamoto S, Oka H, Kimura Y, Tamura S, Inoue M, Monden T, Ito F & Monden M (1996) Beta-catenin expression in human cancers. Am J Pathol 148(1): 39–46.

Takeichi M, Hatta K, Nose A & Nagafuchi A (1988) Identification of a gene family of cadherin cell adhesion molecules. Cell Differ Dev 25 Suppl: 91–94.

Tamura S, Shiozaki H, Miyata M, Kadowaki T, Inoue M, Matsui S, Iwazawa T, Takayama T, Takeichi M & Monden M (1996) Decreased E-cadherin expression is associated with haematogenous recurrence and poor prognosis in patients with squamous cell carcinoma of the oesophagus. Br J Surg 83(11): 1608–1614.

Tanaka J, Sugimoto K, Shiraki K, Tameda M, Kusagawa S, Nojiri K, Beppu T, Yoneda K, Yamamoto N, Uchida K, Kojima T & Takei Y (2010) Functional cell surface expression of toll-like receptor 9 promotes cell proliferation and survival in human hepatocellular carcinomas. Int J Oncol 37(4): 805–814.

Taraif SH, Deavers MT, Malpica A & Silva EG (2009) The Significance of Neuroendocrine Expression in Undifferentiated Carcinoma of the Endometrium. Int J Gynecol Pathol 28(2): 142–147.

Taupenot L, Harper KL & O'Connor DT (2003) The chromogranin-secretogranin family. N Engl J Med 348(12): 1134–1149.

Terpe HJ, Tajrobehkar K, Günthert U & Altmannsberger M (1993) Expression of cell adhesion molecules alpha-2, alpha-5 and alpha-6 integrin, E-cadherin, N-CAM and CD-44 in renal cell carcinomas. An immunohistochemical study. Virchows Arch A Pathol Anat Histopathol 422(3): 219–224.

Thompson JA, Kuzel T, Drucker BJ, Urba WJ & Bukowski RM (2009) Safety and efficacy of PF-3512676 for the treatment of stage IV renal cell carcinoma: an open-label, multicenter phase I/II study. Clin Genitourin Cancer 7(3): E58–65.

Thompson RH, Kuntz SM, Leibovich BC, Dong H, Lohse CM, Webster WS, Sengupta S, Frank I, Parker AS, Zincke H, Blute ML, Sebo TJ, Cheville JC & Kwon ED (2006) Tumor B7-H1 is associated with poor prognosis in renal cell carcinoma patients with long-term follow-up. Cancer Res 66(7): 3381–3385.

Thun MJ, Namboodiri MM & Heath CW Jr (1991) Aspirin use and reduced risk of fatal colon cancer. N Engl J Med 325(23): 1593–1596.

131

Tokunaga T, Yamamoto H, Shimada S, Abe H, Fukuda T, Fujisawa Y, Furutani Y, Yano O, Kataoka T & Sudo T (1984) Antitumor activity of deoxyribonucleic acid fraction from Mycobacterium bovis BCG. I. Isolation, physicochemical characterization, and antitumor activity. J Natl Cancer Inst 72(4): 955–962.

Tomchuck SL, Zwezdaryk KJ, Coffelt SB, Waterman RS, Danka ES & Scandurro AB (2008) Toll-like receptors on human mesenchymal stem cells drive their migration and immunomodulating responses. Stem Cells 26(1): 99–107.

Toyokuni S, Okamoto K, Yodoi J & Hiai H (1995) Persistent oxidative stress in cancer. FEBS Lett 358(1): 1–3.

Trachootham D, Alexandre J & Huang P (2009) Targeting cancer cells by ROS-mediated mechanisms: a radical therapeutic approach? Nat Rev Drug Discov 8(7): 579–591.

Tsui KH, Shvarts O, Smith RB, Figlin RA, deKernion JB & Belldegrun A (2000) Prognostic indicators for renal cell carcinoma: a multivariate analysis of 643 patients using the revised 1997 TNM staging criteria. J Urol 163(4): 1090–1095; quiz 1295.

Tsujii M & DuBois RN (1995) Alterations in cellular adhesion and apoptosis in epithelial cells overexpressing prostaglandin endoperoxide synthase 2. Cell 83(3): 493–501.

Tsujii M, Kawano S & DuBois RN (1997) Cyclooxygenase-2 expression in human colon cancer cells increases metastatic potential. Proc Natl Acad Sci U S A 94(7): 3336–3340.

Tsujii M, Kawano S, Tsuji S, Sawaoka H, Hori M & DuBois RN (1998) Cyclooxygenase regulates angiogenesis induced by colon cancer cells. Cell 93(5): 705–716.

Tuna B, Yorukoglu K, Gurel D, Mungan U & Kirkali Z (2004) Significance of COX-2 expression in human renal cell carcinoma. Urology 64(6): 1116–1120.

Turunen N, Karihtala P, Mäntyniemi A, Sormunen R, Holmgren A, Kinnula VL & Soini Y (2004) Thioredoxin is associated with proliferation, p53 expression and negative estrogen and progesterone receptor status in breast carcinoma. APMIS 112(2): 123–132.

Tutton PJ & Barkla DH (1987) Biogenic amines as regulators of the proliferative activity of normal and neoplastic intestinal epithelial cells (review). Anticancer Res 7(1): 1–12.

Uzzo RG & Novick AC (2001) Nephron sparing surgery for renal tumors: indications, techniques and outcomes. J Urol 166(1): 6–18.

Uzzo RG, Rayman P, Kolenko V, Clark PE, Bloom T, Ward AM, Molto L, Tannenbaum C, Worford LJ, Bukowski R, Tubbs R, Hsi ED, Bander NH, Novick AC & Finke JH (1999) Mechanisms of apoptosis in T cells from patients with renal cell carcinoma. Clin Cancer Res 5(5): 1219–1229.

Väisänen MR, Väisänen T, Jukkola-Vuorinen A, Vuopala KS, Desmond R, Selander KS & Vaarala MH (2010) Expression of toll-like receptor-9 is increased in poorly differentiated prostate tumors. Prostate 70(8): 817–824.

van der Poel HG, Roukema JA, Horenblas S, van Geel AN & Debruyne FM (1999) Metastasectomy in renal cell carcinoma: A multicenter retrospective analysis. Eur Urol 35(3): 197–203.

132

van Kilsdonk JW, van Kempen LC, van Muijen GN, Ruiter DJ & Swart GW (2010) Soluble adhesion molecules in human cancers: sources and fates. Eur J Cell Biol 89(6): 415–427.

Van Poppel H, Da Pozzo L, Albrecht W, Matveev V, Bono A, Borkowski A, Colombel M, Klotz L, Skinner E, Keane T, Marreaud S, Collette S & Sylvester R (2011) A prospective, randomised EORTC intergroup phase 3 study comparing the oncologic outcome of elective nephron-sparing surgery and radical nephrectomy for low-stage renal cell carcinoma. Eur Urol 59(4): 543–552.

Vane JR (1971) Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nat New Biol 231(25): 232–235.

Vane JR, Bakhle YS & Botting RM (1998) Cyclooxygenases 1 and 2. Annu Rev Pharmacol Toxicol 38: 97–120.

Viallet J & Ihde DC (1991) Small cell carcinoma of the lung: clinical and biologic aspects. Crit Rev Oncol Hematol 11(2): 109–135.

Vicaut E, Laemmel E & Stücker O (2000) Impact of serotonin on tumour growth. Ann Med 32(3): 187–194.

Vinores SA, Bonnin JM, Rubinstein LJ & Marangos PJ (1984) Immunohistochemical demonstration of neuron-specific enolase in neoplasms of the CNS and other tissues. Arch Pathol Lab Med 108(7): 536–540.

Vogelstein B & Kinzler KW (2004) Cancer genes and the pathways they control. Nat Med 10(8): 789–799.

Vogelzang NJ, Priest ER & Borden L (1992) Spontaneous regression of histologically proved pulmonary metastases from renal cell carcinoma: a case with 5-year followup. J Urol 148(4): 1247–1248.

Volpe A & Patard JJ (2010) Prognostic factors in renal cell carcinoma. World J Urol 28(3): 319–327.

Vreugde S, Ferrai C, Miluzio A, Hauben E, Marchisio PC, Crippa MP, Bussi M & Biffo S (2006) Nuclear myosin VI enhances RNA polymerase II-dependent transcription. Mol Cell 23(5): 749–755.

Wagner H (2004) The immunobiology of the TLR9 subfamily. Trends Immunol 25(7): 381–386.

Wakabayashi N, Slocum SL, Skoko JJ, Shin S & Kensler TW (2010) When NRF2 talks, who's listening? Antioxid Redox Signal 13(11): 1649–1663.

Wang C, Cao S, Yan Y, Ying Q, Jiang T, Xu K & Wu A (2010) TLR9 expression in glioma tissues correlated to glioma progression and the prognosis of GBM patients. BMC Cancer 10: 415.

Wang W, Caldwell MC, Lin S, Furneaux H & Gorospe M (2000) HuR regulates cyclin A and cyclin B1 mRNA stability during cell proliferation. EMBO J 19(10): 2340–2350.

Wang W, Fan J, Yang X, Fürer-Galbán S, López de Silanes I, von Kobbe C, Guo J, Georas SN, Foufelle F, Hardie DG, Carling D & Gorospe M (2002) AMP-activated kinase regulates cytoplasmic HuR. Mol Cell Biol 22(10): 3425–3436.

133

Wang W, Yang X, Cristofalo VJ, Holbrook NJ & Gorospe M (2001) Loss of HuR is linked to reduced expression of proliferative genes during replicative senescence. Mol Cell Biol 21(17): 5889–5898.

Warner CL, Stewart A, Luzio JP, Steel KP, Libby RT, Kendrick-Jones J & Buss F (2003) Loss of myosin VI reduces secretion and the size of the Golgi in fibroblasts from Snell's waltzer mice. EMBO J 22(3): 569–579.

Weiss RH, Borowsky AD, Seligson D, Lin PY, Dillard-Telm L, Belldegrun AS, Figlin RA & Pantuck AD (2007) P21 is a Prognostic Marker for Renal Cell Carcinoma: Implications for Novel Therapeutic Approaches. J Urol 177(1): 63–68; discussion 68–69.

Wells AL, Lin AW, Chen LQ, Safer D, Cain SM, Hasson T, Carragher BO, Milligan RA & Sweeney HL (1999) Myosin VI is an actin-based motor that moves backwards. Nature 401(6752): 505–508.

Westlin J, Edgren M, Letocha H, Stridsberg M, Wilander E & Nilsson S (1995) Positron emission tomography utilizing C-11 5-hydroxytryptophan, plasma biochemistry and neuroendocrine immunohistochemistry of metastatic renal-cell carcinoma. Oncol Rep 2(4): 543–548.

Wiedenmann B & Huttner WB (1989) Synaptophysin and chromogranins/secretogranins--widespread constituents of distinct types of neuroendocrine vesicles and new tools in tumor diagnosis. Virchows Arch B Cell Pathol Incl Mol Pathol 58(2): 95–121.

Williams CS, Mann M & DuBois RN (1999) The role of cyclooxygenases in inflammation, cancer, and development. Oncogene 18(55): 7908–7916.

Wiseman H & Halliwell B (1996) Damage to DNA by reactive oxygen and nitrogen species: role in inflammatory disease and progression to cancer. Biochem J 313 (Pt 1): 17–29.

Yamamoto T, Suzuki T, Kobayashi A, Wakabayashi J, Maher J, Motohashi H & Yamamoto M (2008) Physiological significance of reactive cysteine residues of Keap1 in determining Nrf2 activity. Mol Cell Biol 28(8): 2758–2770.

Yaman O, Baltaci S, Arikan N, Ozdiler E, Gogus O & Muftuoglu YZ (1996) Serum neuron specific enolase: can it be a tumour marker for renal cell carcinoma? Int Urol Nephrol 28(2): 207–210.

Yao M, Yoshida M, Kishida T, Nakaigawa N, Baba M, Kobayashi K, Miura T, Moriyama M, Nagashima Y, Nakatani Y, Kubota Y & Kondo K (2002) VHL tumor suppressor gene alterations associated with good prognosis in sporadic clear-cell renal carcinoma. J Natl Cancer Inst 94(20): 1569–1575.

Yi X, Zhou Y, Zheng W & Chambers SK (2009) HuR expression in the nucleus correlates with high histological grade and poor disease-free survival in ovarian cancer. Aust N Z J Obstet Gynaecol 49(1): 93–98.

Yoshida H, Cheng W, Hung J, Montell D, Geisbrecht E, Rosen D, Liu J & Naora H (2004) Lessons from border cell migration in the Drosophila ovary: A role for myosin VI in dissemination of human ovarian cancer. Proc Natl Acad Sci U S A 101(21): 8144–8149.

134

Yoshimura R, Matsuyama M, Kawahito Y, Tsuchida K, Kuratsukuri K, Takemoto Y, Mitsuhashi M, Sano H & Nakatani T (2004) Study of cyclooxygenase-2 in renal cell carcinoma. Int J Mol Med 13(2): 229–233.

Youssef YM, White NM, Grigull J, Krizova A, Samy C, Mejia-Guerrero S, Evans A & Yousef GM (2011) Accurate molecular classification of kidney cancer subtypes using microRNA signature. Eur Urol 59(5): 721–730.

Yuen JS, Cockman ME, Sullivan M, Protheroe A, Turner GD, Roberts IS, Pugh CW, Werner H & Macaulay VM (2007) The VHL tumor suppressor inhibits expression of the IGF1R and its loss induces IGF1R upregulation in human clear cell renal carcinoma. Oncogene 26(45): 6499–6508.

Zhu X, Kanai Y, Saito A, Kondo Y & Hirohashi S (2000) Aberrant expression of beta-catenin and mutation of exon 3 of the beta-catenin gene in renal and urothelial carcinomas. Pathol Int 50(12): 945–952.

Zifa E & Fillion G (1992) 5-Hydroxytryptamine receptors. Pharmacol Rev 44(3): 401–458. Zisman A, Pantuck AJ, Dorey F, Said JW, Shvarts O, Quintana D, Gitlitz BJ, deKernion

JB, Figlin RA & Belldegrun AS (2001) Improved prognostication of renal cell carcinoma using an integrated staging system. J Clin Oncol 19(6): 1649–1657.

135

Original publications

I Ronkainen H, Vaarala MH, Kauppila S, Soini Y, Paavonen TK, Rask J & Hirvikoski P (2009) Increased BTB-Kelch type substrate adaptor protein immunoreactivity associates with advanced stage and poor differentiation in renal cell carcinoma. Oncol Rep 21(6): 1519–1523.

II Ronkainen H, Kauppila S, Hirvikoski P & Vaarala MH (2010) Evaluation of myosin VI, E-cadherin and beta-catenin immunostaining in renal cell carcinoma. J Exp Clin Cancer Res 29: 2.

III Ronkainen H, Vaarala MH, Hirvikoski P & Ristimäki A (2011) HuR expression is a marker of poor prognosis in renal cell carcinoma. Tumor Biol 32(3): 481–487.

IV Ronkainen H, Hirvikoski P, Kauppila S, Vuopala KS, Paavonen TK & Selander KS, Vaarala MH (2011) Absent Toll-like receptor 9 expression predicts poor prognosis in renal cell carcinoma. J Exp Clin Cancer Res 30: 84.

V Ronkainen H, Soini Y, Vaarala MH, Kauppila S & Hirvikoski P (2010) Evaluation of neuroendocrine markers in renal cell carcinoma. Diagn Pathol 5: 28.

Reprinted with permission from Spandidos Publications (I), BioMed Central (II,

IV, V) and Springer (III).

Original publications are not included in the electronic version of the dissertation.

136

A C T A U N I V E R S I T A T I S O U L U E N S I S

Book orders:Granum: Virtual book storehttp://granum.uta.fi/granum/

S E R I E S D M E D I C A

1133. Paananen, Markus (2011) Multi-site musculoskeletal pain in adolescence:occurrence, determinants, and consequences

1134. Kaustinen, Teija (2011) Oulu-hoitoisuusluokitus ja hoitohenkilökunnan ajankäyttöhoitotyön laatuvaatimusten näkökulmasta

1135. Liukkonen, Timo (2011) Low-grade inflammation in depression, anxiety and sleepdisturbances

1136. Tölli, Hanna (2011) Reindeer-derived bone protein extract in the healing of bonedefects : Evaluation of various carrier materials and delivery systems

1137. Tourula, Marjo (2011) The childcare practice of children’s daytime sleepingoutdoors in the context of Northern Finnish winter

1138. Mäkelä, Jussi (2011) Bone marrow-derived stem cell therapy in acute myocardialinfarction : An experimental porcine model

1139. Törmänen, Outi (2011) Malli kunnallisten terveyspalveluiden arvokeskustelusta :Pehmeä systeemianalyysi kolmen kunnan yhteistoiminta-alueella

1140. Kangas, Maarit (2011) Development of accelerometry-based fall detection : fromlaboratory environment to real life

1141. Määttä, Tuomo (2011) Down syndrome, health and disability : A population-based case record and follow-up study

1142. Leskelä, Tarja (2011) Human δ opioid receptor Phe27 and Cys27 variants : Therole of heteromerization and pharmacological chaperones in receptor processingand trafficking

1143. Karjalainen, Minna (2011) Genetic predisposition to spontaneous preterm birth :approaches to identify susceptibility genes

1144. Saaristo, Timo (2011) Assessment of risk and prevention of type 2 diabetes inprimary health care

1145. Vuononvirta, Tiina (2011) Etäterveydenhuollon käyttöönotto terveydenhuollonverkostoissa

1146. Vanhala, Marja (2012) Lapsen ylipaino – riskitekijät, tunnistaminen ja elintavat

1147. Katisko, Jani (2012) Intraoperative imaging guided delineation and localization ofregions of surgical interest : Feasibility study

1148. Holmström, Anneli (2012) Etnografinen tutkimus natiivitutkimusten oppimisestaröntgenhoitajaopiskelijoiden opinnoissa

ABCDEFG

UNIVERS ITY OF OULU P.O.B . 7500 F I -90014 UNIVERS ITY OF OULU F INLAND

A C T A U N I V E R S I T A T I S O U L U E N S I S

S E R I E S E D I T O R S

SCIENTIAE RERUM NATURALIUM

HUMANIORA

TECHNICA

MEDICA

SCIENTIAE RERUM SOCIALIUM

SCRIPTA ACADEMICA

OECONOMICA

EDITOR IN CHIEF

PUBLICATIONS EDITOR

Senior Assistant Jorma Arhippainen

Lecturer Santeri Palviainen

Professor Hannu Heusala

Professor Olli Vuolteenaho

Senior Researcher Eila Estola

Director Sinikka Eskelinen

Professor Jari Juga

Professor Olli Vuolteenaho

Publications Editor Kirsti Nurkkala

ISBN 978-951-42-9772-4 (Paperback)ISBN 978-951-42-9773-1 (PDF)ISSN 0355-3221 (Print)ISSN 1796-2234 (Online)

U N I V E R S I TAT I S O U L U E N S I S

MEDICA

ACTAD

D 1149

ACTA

Hanna-Leena R

onkainen

OULU 2012

D 1149

Hanna-Leena Ronkainen

NOVEL PROGNOSTIC BIOMARKERS FOR RENAL CELL CARCINOMA

UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU, FACULTY OF MEDICINE,INSTITUTE OF CLINICAL MEDICINE, DEPARTMENT OF SURGERY;INSTITUTE OF DIAGNOSTICS, DEPARTMENT OF PATHOLOGY