· Stellingen behorende bij het proefschrift Molecular and biological nature of endometrial 1. Om de prognose van patienten met een uterien carcinosarcoom te verbeteren, toekomstige

University of Groningen

Molecular and biological nature of endometrial cancerJong, Renske Akke de

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2012

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Jong, R. A. D. (2012). Molecular and biological nature of endometrial cancer. [s.n.].

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 21-04-2021

https://research.rug.nl/en/publications/molecular-and-biological-nature-of-endometrial-cancer(7356e8f5-b0cb-4939-881d-f37759723357).html

Molecular and biological nature of

\ endometrial � # cancer

Renske A. de Jong

Molecular and biological nature of endometrial cancer

Renske A. de Jong

Stellingen behorende bij het proefschrift

Molecular and biological nature of endometrial

1. Om de prognose van patienten met een uterien carcinosarcoom te verbeteren, toekomstige behandelingen zich te richten op histologische eigenschappen v epitheliale tumorcomponent. (dit proefschrift)

Ci.·ntru!e

\1:::d.�che

�g�ifthcek romngen

2. De slechtere overleving van patienten met afwijkende DNA-reparatie-eiwitten is ten dele te wijten aan het feit dat het HLA klasse I molecuul vaker is afgeschakeld waardoor cytotoxische T-cellen de kankercellen niet kunnen herkennen en doden. (dit proefschrift)

3. Patienten met endometriumcarcinoom bij wie de tumor afwijkende expressie van DNAreparatie-eiwitten vertoont, zijn minder gebaat bij immunotherapie clan patienten met normale expressie van DNA-reparatie-eiwitten in de tumor. (dit proefschrift)

4. Bij patienten met een (gynaecologische) maligniteit is de enzymatische afbraak van tryptofaan via de kynurenine pathway verhoogd. (dit proefschrift)

5. Blokkade van indoleamine 2,3-dioxygenase is een nieuwe, mogelijk veelbelovende, aanpak voor de verbetering van de prognose van patienten met endometriumcarcinoom. (o.a. dit proefschrift)

6. Hoewel radiotherapie zorgt voor uitstekende lokale controle, leidt dit niet tot een verbeterde overleving van patienten met endometriumcarcinoom. (Lancet 2009;373:137-146)

7. Verkleining van het uitwendige bestralingsveld leidt tot vermindering van gastrointestinale bijwerkingen bij patienten met endometriumcarcinoom. (dit proefschrift)

8. "No evidence" doesn't mean "no efficacy". (vrij naar: Gillian M. Thomas)

9. Variability is the law of life, and as no two faces are the same, so no bodies are alike, and no two individuals react alike and behave alike under the abnormal condition which we know as disease. (William Osler)

10. We can't solve problems by using the same kind of thinking we used when we created them. (Albert Einstein)

11. Tijdens de opleiding tot medisch specialist zou meer aandacht besteed moeten warden aan pathologieonderwijs.

12. I don't sing because I'm happy; I'm happy because I sing. (William James)

Renske A. de Jong Groningen, 7 november 2012

u

M

C

G

De Jong, R.A. Molecular and biological nature of endometrial cancer

Thesis, University of Groningen, the Netherlands

© Copyright 2012 Renske A. de Jong, Groningen, the Netherlands All rights reserved. No part of this thesis may be reproduced, stored in a retrieval system, or transmitted in any form or by any means without prior permission of the author and the publisher holding the copyright of the published articles.

Cover design and lay-out: Ankemarije Dam: www.ankemarijedam.nl Printed by: Ipskamp Drukkers, Enschede, the Netherlands

The research presented in this thesis was financially supported by the Jan Kornelis de Cock-Stichting, University Medical Center Groningen, the Netherlands.

Printing of this thesis was financially supported by:

Graduate School for Drug Exploration (GUIDE) Intrada Muziekuitgave MediReva Sanofi Pasteur MSD

-Mundipharma Pharmaceuticals B.V.

Roche Nederland B.V Therabel Pharma B.V. University of Groningen University Medical Center Groningen

0 ELEKTA

RIJKSUNIVERSITEIT GRONINGEN

Molecular and biological nature of endometrial cancer

Proefschrift

ter verkrijging van het doctoraat in de

Medische wetenschappen·

aan de Rijksuniversiteit Groningen

op gezag van de

Rector Magnificus, dr. E. Sterken,

in het openbaar te verdedigen op

woensdag 7 november 2012

om 14.30 uur

door

Renske Akke de Jong

geboren op 5 juli 1984

te Heerenveen

Ccntrale u Medische M Bibliothcek C

Groningcn G

t��-

Prornotores: Prof. dr. H.W. Nijrnan

Prof. dr. H. Hollerna

Prof. dr. LP. Kerna

Beoordelingscornrnissie: Prof. dr. F. Arnant

Prof. dr. J.A. Langendijk

Prof. dr. L.F.M.H. de Leij

Foar heit en mem

Contents

Chapter 1 General introduction and thesis outline 9

Chapter 2 Presence of tumor-infiltrating lymphocytes is an independent 21 prognostic factor in Type I and II endometrial cancer Gynecologic Oncology 2009;114:105-110.

Chapter 3 Loss of HLA class I and mismatch repair protein expression in 37 sporadic endometrioid endometrial carcinomas International Journal of Cancer 2012;131:1828-36.

Chapter 4 Serum tryptophan and kynurenine concentrations as parameters 55 for indoleamine 2,3-dioxygenase activity in patients with endometrial, ovarian and vulvar cancer International Journal of Gynecological Cancer 2011;21:1320-1327.

Chapter 5 Immunosuppressive mechanism of indoleamine 2,3-dioxygenase 69 and its prognostic role in endometrial carcinoma Partly based on: Prognostic role of indoleamine 2,3-dioxygenase in endometrial carcinoma. Gynecologic Oncology 2012;126:474-80.

Chapter 6 Molecular markers and clinical behavior of uterine 89 carcinosarcomas: focus on the epithelial tumor component Modern Pathology 2011;24:1368-1379.

Chapter 7 Less gastrointestinal toxicity after adjuvant radiotherapy on a 111 small pelvic field compared to a standard pelvic field in patients with endometrial carcinoma International Journal of Gynecological Cancer 2012;22:1177-86.

Chapters Summarizing discussion and future perspectives 129

Chapter 9 Nederlandse samenvatting 145

Abbreviations 155

List of co-authors 159

Dankwoord 163

Publication overview 169

Curriculum Vitae 173

Chapter 1

General introduction and thesis outline


General introduction

This thesis aimed to increase knowledge of the biological and clinical behavior of endometrial carcinoma (EC) in order to improve treatment and prognosis in the future. The role of the immune system in EC was especially emphasized.

This chapter is an introduction to EC and describes pathological features of EC subtypes and standard and innovative treatment strategies. Immunotherapy is currently under investigation in several types of cancer including EC where it is generally recognized that the immune system plays an important role in regulating cancer development and progression. Therefore, general aspects of the immune system are discussed in this chapter as well.

Endometrial carcinoma EC refers to a malignancy that arises from the endometrium of the uterus and more specifically, from epithelial cells that line the endometrium and form the endometrial glands. EC is the most common gynecological malignancy and the fourth most common malignancy in women in the western world. Based on clinicopathological characteristics, EC can be divided in two different types; type I consists of low-grade tumors with an endometrioid tumor type whereas type II EC consists of non-endometrioid tumor types (clearcell and serous carcinoma).1-2 Type I EC represents 80% of all EC cases and arises on the background of hyperplasia after unopposed estrogen stimulation. Mutations of DNA mismatch repair (MMR) genes, PTEN mutations, estrogen receptor expression, progesterone receptor expression and aberrant Wnt/l3-catenin signaling pathway occur frequently amongst type I EC. Type II tumors are not estrogen driven and occasionally arise in endometrial polyps or from precancerous lesions ( endometrial intraepithelial hyperplasia) in the presence of atrophic endometrium. Type II EC are characterized by p53 mutations and have a worse prognosis compared to type I.3 Postmenopausal bleeding is an early presenting symptom of EC and therefore most patients (75%) are diagnosed with early stage of disease. Overall survival is good in early stage EC; 87% for stage I and 76% for stage II but declines with advanced disease (stage III: 57%, stage IV: 18%).4 Extension of EC occurs by direct invasion of the myometrium and the cervix. The most common route of spreading outside the uterus is through the lymphatic vessels; locoregional pelvic lymph nodes are most commonly affected whereas para-aortic metastases are less common.5•6 The most important prognostic factors irt EC are stage of disease, myometrial invasion, histological type, tumor grade and lymphvascular space involvement (LVSI).3•7

Uterine carcinosarcoma; metaplastic subtype of EC Uterine carcinosarcoma (UCS) is an aggressive subtype of EC and accounts for only 1-5% of uterine malignancies. UCS consist of two histological malignant components; an epithelial (carcinoma) and a mesenchymal (sarcoma) component.8•

9 The coexistence of two distinctive malignant components in the same tumor was extensively investigated and is now explained by the generally accepted 'conversion theory'; the epithelial

11

Chapter 1

12

component is the 'driving force' and the mesenchymal component is derived from the epithelial component via a metaplastic process.10 UCS has a high tendency to early extrauterine spread and therefore, advanced disease is usually present at time of diagnosis. Prognosis is poor; 5-year survival rates have been reported between 30% and 45.8% in early stage UCS (stage I/II) and 0% to 10% in advanced stage UCS (stage III/IV).11·12

Staging and treatment In 1988, the International Federation of Gynecology and Obstetrics (FIGO) introduced a surgical staging system for EC which has been revised in 2009 (Table 1 and 2). Clinicopathological data and tissue material described in this thesis originate from a time period prior to 2009. Therefore, when stage of disease is depicted, this refers to the 1988 FIGO staging system. Surgery is the cornerstone of EC treatment and consists of hysterectomy and bilateral salpingo-oopherectomy (BSO) either by laparotomy or laparoscopy (in clinically stage I EC)_B,14

The FIGO originally recommended complete dissection of pelvic and para-aortic lymph nodes as part of the surgical procedure. However, the role of lymphadenectomy has been one of the most controversial aspects in the surgical management of EC over the last years. Based on large and randomized controlled trials, it is now generally agreed that lymphadenectomy can be omitted in low-risk EC.15•16 Where patients with high-intermediate risk EC have an increased risk for lymph node metastases, distant metastases and death due to EC, adequate surgical staging including lymphadenectomy is helpful to determine prognosis and adjuvant treatment strategies.17•18

Patients with low-risk EC are treated with surgery alone. For patients with intermediate-risk, surgery is followed by radiotherapy if 2 out of 3 risk factors are present; age over 60 years, �50% myometrial invasion and tumor grade 3. 19 Adjuvant radiotherapy is always indicated for patients with high-risk EC. Although the use of adjuvant radiotherapy in EC reduces the rate of locoregional recurrences, no benefit for survival has been observed so far.19-21 Especially high-risk EC patients have increased risk for distant metastases and death due to EC, and trials investigating the role of adjuvant chemotherapy in high-risk EC are currently ongoing.22

Over the last few years, not much progress has been made in improving prognosis of EC patients. This especially accounts for patients with an aggressive subtype (type II EC or UCS). It is therefore needed to further characterize these subtypes and explore new treatment strategies. In this respect, modulating the anti-cancer immune response in EC is a logical option because the cancer immune response itself has been a subject of interest in several cancer types. Reports in EC have focused on defining prognostic factors of the immune system and immunological targets for immunotherapy.23-26

Tumor immunology and immune escape It is generally recognized that the immune system is important in regulating cancer development. First, Burnet and Thomas described the "cancer immunosurveillance" theory in which the immune system can recognize and kill cancer cells.27

-29 Nowadays,

the concept has been extended to "cancer immunoediting" which includes the cancer immunosurveillance theory but recognizes that the immune system can also function


to promote or select tumor variants with reduced immunogenicity.30•31

As part of the adaptive immune system, cytotoxic T-lymphocytes (CTL) are subtypes of T-lymphocytes able to recognize and subsequently kill cells presenting peptides of "non-self" origin and thus protect the host from foreign pathogens such as viruses and cancer. CTLs express T-cell receptors (TCR) which can recognize specific antigenic peptides derived from digested proteins which are broken down in the proteasome of the cell and transported to the cell surface. Recognition of "non-self" can only take place when antigenic peptides are presented by the human leukocyte antigen (HLA) class I (or: major histocompatibility complex, MHC) present at the cell surface of every nucleated cell. HLA class I comprises the classical (class Ia) HLA-A, -B and -C and the nonclassical ( class lb) HLA-E, -F and -G proteins (heavy a chain) and a non polymorphic �

2 -

microglobulin (�2m) light chain. The importance of CTL as part of anti-tumor response was emphasized by studies in several cancers demonstrating that a high number of tumor-infiltrating CDS+ T-lymphocytes (CTL) is associated with increased survival.26•32-34

However, cancers are able to develop mechanisms to escape from a tumor-specific immune response.30 Downregulation of the HLA class I molecule is one such powerful mechanism, preventing immune recognition and lysis by CTLs. Next, the intracellular

Table 1. 1988 FICO staging system for endometrial carcinoma

Stage

IA IB IC IIA IIB IIIA IIIB me IVA IVB

Criteria

Tumor limited to the endometrium Tumor invades < 50% myometrium Tumor invades > 50% myometrium Extension to cervical mucosa Extension to cervical stroma Extension to uterine serosa, adnexae or positive cytology Vaginal metastases Metastasis to pelvic and/or para-aortic lymph nodes Extension to bladder and/or rectal mucosa Distant metastasis including intra-abdominal and/or inguinal lymph nodes

Table 2. 2009 FICO staging system for endometrial carcinoma

Stage

IA IB II IIIA IIIB IIICl IIIC2 IVA IVB

Criteria

Tumor limited to the endometrium or invades < 50% myometrium Tumor invades > 50% myometrium Extension to cervical stroma Extension to uterine serosa and/or adnexae Extension to vagina, parametria and/or pelvic peritoneum Metastasis to pelvic lymph nodes Metastasis to para-aortic lymph nodes regarless of pelvic node status Extension to bladder and/or rectal mucosa Distant metastasis including intra-abdominal and/or inguinal lymph nodes

13

Chapter 1

14

enzyme indoleamine 2,3-dioxygenase (IDO; EC 1.13.11.52) is currently considered as an important immune escape mechanisms in cancer and has been associated with worse prognosis in several cancer types.35-39 IDO catalyses the first and rate-limiting steps in the breakdown of the essential amino acid tryptophan along the kynurenine pathway where several downstream metabolites such as kynurenine, 3-hydroxykynurenine and ultimately nicotinamide adenine dinucleotide (NAD+) are formed (Figure 1). IDO is mainly induced by IFN-y, which is produced by natural killer (NK) cells and CTLs. Next to IFN-y, the release of lipopolysaccharide can elicit IDO expression in several cell types (e.g. dendritic cells or cancer cells). Originally, the immunosuppressive potential of IDO was described in murine pregnancies where it prevents allogeneic fetal rejection by maternal T-cell immunity.40 IDO exerts its immunosuppressive function by suppressing effector T-lymphocytes and NK cells. A few mechanisms have been proposed. First, local tryptophan depletion leads to cell cycle arrest in the Gl phase and increased susceptibility for apoptosis of T-lymphocytes.41 Second, metabolites are released due to tryptophan breakdown which are directly toxic to cytotoxic T-lymphocytes and NK cells.4244 The third mode of immune suppressive action is the ability of IDO to convert nai:ve T-cells to differentiate into Foxp3+ regulatory T-cells (Treg) which are well-known for their immunosuppressive potential.45

Figure 1. Role of JOO in the kynurenine pathway; breakdown of tryptophan (simplified scheme). NAO+: nicotinamide adenine dinucleotide.

Tryptophan @)1�

Kynurenine Kynurenic acid � l

3-Hydroxykynurenine 1 Quino linic acid 1

NAD+


Outline of this thesis

The aim of chapter 2 was to determine the presence and the prognostic significance of three important subtypes of T-lymphocytes in EC. Therefore, numbers of intra-tumoral CD8+ cytotoxic T-lymphocytes, CD45RQ+ memory T-lymphocytes and Foxp3+ regulatory T-lymphocytes were determined by immunohistochemistry in primary tumor tissue of 368 EC patients. Next, type I and type II EC were analyzed separately in order to determine the role of the immune system in both tumor types.

Immunotherapy is based on the mechanism of inducing a tumor-specific cytotoxic immune response. However, when HLA class I is downregulated, antigen presentation to CTLs is impaired and therefore, tumors can escape recognition and destruction by cells of the adaptive immune system. It is important to understand in which tumors HLA class I downregulation occurs more frequently in order to adjust treatment strategies. For example, in colorectal carcinomas with deficiencies of the DNA mismatch repair (MMR) system, HLA class I downregulation more frequently occurred compared to tumors with proficient MMR system.46 Chapter 3 investigated whether such an association is also present in EC. Therefore, immunostaining of three important MMR proteins (MLHl, MSH2 and MSH6) and HLA class I downregulation were determined in a cohort of 486 endometrioid EC patients. The influence of the MMR protein expression status on the immune system was further explored by determining if intra-tumoral CDS+

T-lymphocytes (as described in chapter 2) were associated to MMR protein expression or HLA class I downregulation and if there was an influence on survival.

HLA class I downregulation is not the sole mechanism by which tumors escape immune responses. Currently, IDO is considered as an important immune escape mechanism. The aim of chapter 4 was to determine if IDO activity can be determined in serum of cancer patients as a reflection of cancer-induced immune escape. Therefore, concentrations of tryptophan and kynurenine (which is the first catabolic product of tryptophan under the influence of IDO) were determined in serum of patients with endometrial, ovarian and vulvar cancer and compared to healthy controls.

The prognostic role of IDO was determined by immunohistochemistry in primary EC tissue and discussed in chapter 5. Despite the fact that tumor cells and T-lymphocytes are present in the same micro-environment, especially T-lymphocytes are negatively affected by !DO-induced local tryptophan depletion. This chapter also discusses the possible mechanism by which tumor cells remain unaffected by local tryptophan depletion in contrast to effector T-lymphocytes.

Treatment of EC needs further improvement and this especially accounts for UCS. Patients diagnosed with this disease have poor prognosis (overall 5-year survival: 33-39%). Due to the fact that UCS is a rare disease, randomized trials are scarce and improvement of therapy and prognosis has failed so far. In order to improve therapeutic strategies in the future, chapter 6 intended to evaluate molecular characteristics and clinical behavior of UCS. To this end, tissue material of 40 UCS patients was evaluated for 11 molecular markers (ER-a, ER-�, PR-A, PR-B, MLHl, MSH2, MSH6, PTEN, p53,

15

Chapter 1

16

�-catenin and cyclin Dl) which are well established in EC. In order to determine whether treatment modalities should focus on characteristics of the epithelial tumor component, the prognostic role of the epithelial component was determined. Therefore, clinicopathological data were compared between patients with an endometrioid epithelial tumor component and a non-endometrioid epithelial tumor component in order to see if, similar to EC, a distinction can be made in type I and type II carcinoma. Next, clinical behavior of UCS was compared to grade 3 endometrioid and non-endometrioid EC.

Patients with high-risk EC are currently treated with adjuvant radiotherapy in order to prevent locoregional recurrences. However, radiotherapy is associated with short- and long-term toxicity and is related to the treatment volume, daily dose and total dose (amongst others). Chapter 7 evaluated a treatment protocol of the University Medical Center Groningen since 1999. Here, EC patients are treated with surgery including lymphadenectomy based on pre-operatively determined factors (grade 3 endometrioid, non-endometrioid tumor types and/ or cervical invasion). Adjuvant external beam radiotherapy (EBRT) is indicated in case of high-risk factors. 19 EBRT on a small pelvic field (including only the central pelvis and proximal vagina) is applied in case of proven negative lymph nodes. Patients with positive lymph nodes or those who had an inadequate lymphadenectorny are treated with EBRT on a standard pelvic field. Extended field EBRT is given in case of proven common iliac or para-aortic lymph node metastases. In order to determine if treatment volume has an effect on toxicity, acute and late toxicity rates were evaluated among 75 EC patients.

Chapter 8 summarizes the research described in previous chapters and provides suggestions for future research concerning EC.

Chapter 9 is a summary of this thesis in Dutch.


References

1. Bokhman JV . Two pathogenetic types of endometrial carcinoma. Gynecol Oncol 1983;15 :10-17.

2. McCluggage WG, Robboy SJ. Mesenchymal uterine tumors and adenomyosis. In: Robboy SJ, Mutter GL, Prat J, Bentley RC, Russell P, Anderson MC, editors. Pathology of the female reproductive tract: Churchill Livingstone Elsevier; 2009. p . 427-456.

3. Amant F, Moerman P, Neven P, Timmerman D, Van Limbergen E, Vergote I . Endometrial cancer. Lancet 2005;366:491-505.

4. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin 2010;60:277-300.

5. Abu-Rustum NR, Gomez JD, Alektiar KM, Soslow RA, Hensley ML, Leitao MM,Jr., et al. The incidence of isolated paraaortic nodal metastasis in surgically staged endometrial cancer patients with negative pelvic lymph nodes. Gynecol Oncol 2009;115:236-238.

6. Cragun JM, Havrilesky LJ, Calingaert B, Synan I , Secord AA, Soper JT, et al. Retrospective analysis of selective lymphadenectomy in apparent early-stage endometrial cancer. J Clin Oncol 2005;23:3668-3675.

7. Briet JM, Hollema H, Reesink N, Aalders JG, Maurits MJ, ten Hoor KA, et al. Lymphvascular space involvement: an independent prognostic factor in endometrial cancer. Gynecol Oncol 2005;96:799-804.

8. Brooks SE, Zhan M, Cote T, Baquet CR. Surveillance, epidemiology, and end results analysis of 2677 cases of uterine sarcoma 1989-1999. Gynecol Oncol 2004;93:204-208.

9. Lax SF. Molecular genetic changes in epithelial, stromal and mixed neoplasms of the endometrium. Pathology 2007;39:46-54.

10. McCluggage WG. Uterine carcinosarcomas (malignant mixed Mullerian tumors) are metaplastic carcinomas. Int J Gynecol Cancer 2002;12:687-690.

11 . Amant F, Cadron I, Fuso L, Berteloot P, de Jonge E, Jacomen G, et al. Endometrial carcinosarcomas have a different prognosis and pattern of spread compared to high-risk epithelial endometrial cancer. Gynecol Oncol 2005;98:274-280.

12. George E, Lillemoe TJ, Twiggs LB, Perrone T. Malignant mixed mullerian tumor versus highgrade endometrial carcinoma and aggressive variants of endometrial carcinoma: a comparative analysis of survival. Int J Gynecol Pathol 1995;14:39-44.

13. Maurits MJ, Bijen CB, Arts HJ, Ter Brugge HG, van der Sijde R, Paulsen L, et al. Safety of laparoscopy versus laparotomy in early-stage endometrial cancer: a randomised trial. Lancet Oncol. 2010;11:763-771 .

14. Vergote I , Amant F, Neven P. Laparoscopic hysterectomy for early endometrial cancer. Lancet Oncol. 2010;11:707-708.

15 . Benedetti Panici P, Basile S, Maneschi F, Alberto Lissoni A, Signorelli M, Scambia G, et al. Systematic pelvic lymphadenectomy vs. no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst 2008;100:1707-1716.

16. Kitchener H, Swart AM, Qian Q, Amos C, Parmar MK. Efficacy of systematic pelvic lymphadenectomy in endometrial cancer (MRC ASTEC trial) : a randomised study. Lancet 2009;373:125-136.

17. Todo Y, Kato H, Kaneuchi M, Watari H, Takeda M, Sakuragi N. Survival effect of para-aortic lymphadenectomy in endometrial cancer (SEPAL study): a retrospective cohort analysis. Lancet 2010;375:1165-1172.

18. Bernardini MQ, Murphy JK. I ssues surrounding lymphadenectomy in the management of endometrial cancer. J Surg Oncol 2009;99:232-241 .

17

Chapter 1

18

19. Creutzberg CL, van Putten WL, Koper PC, Lybeert ML, Jobsen JJ, Warlam-Rodenhuis CC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355:1404-1411.

20. Blake P, Swart AM, Orton J, Kitchener H, Whelan T, Lukka H, et al. Adjuvant external beam radiotherapy in the treatment of endometrial cancer (MRC ASTEC and NCIC CTG EN.5 randomised trials): pooled trial results, systematic review, and meta-analysis. Lancet 2009;373:137-146.

21. Keys HM, Roberts JA, Brunetto VL, Zaino RJ, Spirtos NM, Bloss JD, et al. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92:744-751 .

22. Creutzberg CL. Adjuvant chemotherapy for endometrial cancer: unproven. Int J Gynecol Cancer 2010;20:560-563.

23. Brooks N, Pouniotis DS. Immunomodulation in endometrial cancer. Int J Gynecol Cancer 2009;19:734-740.

24. Bellati F, Visconti V, Napoletano C, Antonilli M, Frati L, Panici PB, et al. Immunology of gynecologic neoplasms: analysis of the prognostic significance of the immune status. Curr.Cancer Drug Targets. 2009;9:541-565.

25. Varughese J, Cocco E, Bellone S, de Leon M, Bellone M, Todeschini P, et al. Uterine serous papillary carcinomas overexpress human trophoblast-cell-surface marker (trop-2) and are highly sensitive to immunotherapy with hRS7, a humanized anti-trop-2 monoclonal antibody. Cancer 2011;117:3163-72.

26. Kondratiev S, Sabo E, Yakirevich E, Lavie 0, Resnick MB. Intratumoral CD8+ T lymphocytes as a prognostic factor of survival in endometrial carcinoma. Clin Cancer Res 2004;10:4450-4456.

27. Burnet FM. Immunological surveillance in neoplasia. Transplant Rev 1971;7:3-25. 28. Burnet M. Immunological Factors in the Process of Carcinogenesis. Br Med Bull 1964;20:154-

158. 29. Burnet M. Cancer; a biological approach. I. The processes of control. Br Med J 1957;1:779-786. 30. Dunn GP, Bruce AT, Ikeda H, Old LJ, Schreiber RD. Cancer immunoediting: from immunosur

veillance to tumor escape. Nat Immunol 2002;3:991-998. 31. Dunn GP, Old LJ, Schreiber RD. The three Es of cancer immunoediting. Annu Rev Immunol

2004;22:329-360. 32. Sato E, Olson SH, Ahn J, Bundy B, Nishikawa H, Qian F, et al. Intraepithelial CD8+ tumor

infiltrating lymphocytes and a high CDS+ /regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proc Natl Acad Sci U S A 2005;102:18538-18543.

33. Naito Y, Saito K, Shiiba K, Ohuchi A, Saigenji K, Nagura H, et al. CD8+ T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer. Cancer Res 1998;58:3491-3494.

34. Leffers N, Gooden MJ, de Jong RA, Hoogeboom BN, ten Hoor KA, Hollema H, et al. Prognostic significance of tumor-infiltrating T-lymphocytes in primary and metastatic lesions of advanced stage ovarian cancer. Cancer Immunol Immunother 2009;58:449-459.

35. Riesenberg R, Weiler C, Spring 0, Eder M, Buchner A, Popp T, et al. Expression of indoleamine 2,3-dioxygenase in tumor endothelial cells correlates with long-term survival of patients with renal cell carcinoma. Clin Cancer Res 2007;13:6993-7002.


36. Brandacher G, Perathoner A, Ladurner R, Schneeberger S, Obrist P, Winkler C, et al. Prognostic value of indolearnine 2,3-dioxygenase expression in colorectal cancer: effect on tumorinfiltrating T cells. Clin Cancer Res 2006;12:1144-1151.

37. Inaba T, Ino K, Kajiyama H, Shibata K, Yamamoto E, Kondo S, et al. Indoleamine 2,3-dioxygenase expression predicts impaired survival of invasive cervical cancer patients treated with radical hysterectomy. Gynecol Oncol 2010;117:423-428.

38. Inaba T, Ino K, Kajiyama H, Yamamoto E, Shibata K, Nawa A, et al. Role of the immunosuppressive enzyme indolearnine 2,3-dioxygenase in the progression of ovarian carcinoma. Gynecol Oncol 2009;115:185-192.

39. Ino K, Yoshida N, Kajiyama H, Shibata K, Yamamoto E, Kidokoro K, et al. Indoleamine 2,3-dioxygenase is a novel prognostic indicator for endometrial cancer. Br J Cancer 2006;95:1555-1561.

40. Munn DH, Zhou M, Attwood JT, Bondarev I, Conway SJ, Marshall B, et al . Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 1998;281:1191-1193.

41. Munn DH, Shafizadeh E, Attwood JT, Bondarev I, Pashine A, Mellor AL. Inhibition of T cell proliferation by macrophage tryptophan catabolism. J Exp Med 1999;189:1363-1372.

42. Della Chiesa M, Carlomagno S, Frumento G, Balsamo M, Cantoni C, Conte R, et al. The tryptophan catabolite L-kynurenine inhibits the surface expression of NKp46- and NKG2D-activating receptors and regulates NK-cell function. Blood 2006;108:4118-4125.

43. Fallarino F, Grohmann U, Vacca C, Bianchi R, Orabona C, Spreca A, et al. T cell apoptosis by tryptophan catabolism. Cell Death Differ 2002;9:1069-1077.

44. Frumento G, Rotondo R, Tonetti M, Damonte G, Benatti U, Ferrara GB. Tryptophan-derived catabolites are responsible for inhibition of T and natural killer cell proliferation induced by indoleamine 2,3-dioxygenase. J Exp Med 2002;196:459-468.

45. Fallarino F, Grohmann U, You S, McGrath BC, Cavener DR, Vacca C, et al. The combined effects of tryptophan starvation and tryptophan catabolites down-regulate T cell receptor zetachain and induce a regulatory phenotype in naive T cells. J Immunol 2006;176:6752-6761 .

46. Dierssen JW, de Miranda NF, Ferrone S, van Puijenbroek M, Comelisse CJ, Fleuren GJ, et al. HNPCC versus sporadic microsatellite-unstable colon cancers follow different routes toward loss of HLA class I expression. BMC Cancer 2007;7:33.

19

Chapter 2

Presence of tumor-infiltrating lymphocytes is an

independent prognostic factor in type I and II

endometrial cancer

Renske A. de Jong

Ninke Leffers

H. Marike Boezen

K.laske A ten Hoar

Ate G.J. van der Zee

Harry Hollema

Hans W. Nijman

Gynecologic Oncology 2009; 114: 105-10.

Presence of tumor-infiltrating lymphocytes is an independent prognostic factor in type I and II endometrial cancer

Abstract

Introduction Presence of tumor-infiltrating lymphocytes (TIL) is of prognostic importance in a variety of malignancies. This study aims to determine the prognostic value of CDB+ cytotoxic T-lymphocytes (CTL), FoxP3+ regulatory T-lymphocytes (Treg) and CD45R0+ memory T-lymphocytes in endometrial cancer.

Methods The number of tumor infiltrating CDB+, FoxP3+, and CD45RQ+ T-lymphocytes was determined by immunohistochemistry on tissue microarrays containing tumor material from 368 FIGO stage I-IV endometrial cancer patients. Results from immunohistochemistry were correlated with clinicopathological parameters and survival.

Results High numbers of intra-tumoral CDS+ T-lymphocytes, a high CD8+ /FoxP3+ ratio and the presence of CD45R0+ T-lymphocytes were strongly associated with well-known favorable prognostic factors in endometrial cancer. Furthermore, high numbers of CD8+

T-lymphocytes and a high CDs+ /FoxP3+ ratio were associated with a better disease free survival (DFS). High numbers of CDs+ T-lymphocytes and the presence of CD45RQ+

T-lymphocytes were associated with a prolonged overall survival (OS). In multivariate analysis, high numbers of CD8+ T-lymphocytes had an independent prognostic impact for overall survival in the entire cohort (HR 0.48, 95% C.I. 0.26-0.89, p=0.019) and in type II endometrial cancer (HR 0.17, 95% C.I. 0.08-0.36, p<0.001). A high CD8+ /FoxP3+

ratio was independently associated with improved survival in type I endometrial cancer (HR 0.44, 95% C.I. 0.23-0.84, p=0.013). CD45RQ+ lymphocytes were an independent factor for improved OS (HR 0.42, 95% C.I. 0.19-0.93, p=0.033).

Conclusion This study shows that presence of TIL is an independent prognostic factor in endometrial cancer and indicates an important role for the immune system in endometrial cancer.

23

Chapter 2

24

Introduction

Endometrial cancer is the most common gynecological malignancy worldwide and the fourth most common malignancy in women in the developed world after breast, lung and colorectal cancers. It has the lowest mortality of all gynecological malignanciesY Due to postmenopausal bleeding, the disease is often detected at an early stage, which is reflected in the relatively high five-year survival rates. Based on certain clinicopathological characteristics, a distinction can be made in two types of endometrial cancer with different prognostic profiles.3 Type I consists of grade 1 and grade 2 endometrioid carcinomas, where as clear-cell, serous papillary and grade 3 endometrioid carcinomas are type II endometrial carcinomas. The latter have a higher relapse rate and a poorer outcome.

The presence of tumor-infiltrating lymphocytes (TIL) correlates with improved prognosis in patients with several types of cancer.4-9 Each T-lymphocyte subset has a unique role in their antitumor response. The presence of tumor-infiltrating cytotoxic T-lymphocytes (CTL) was linked to an improved survival in endometrial, ovarian and colorectal cancer.7,10,11 Regulatory T-cells (Treg), representing 5-10% of peripheral CD4+

T-lymphocytes12, play an essential role in controlling immune responses to self and non-self antigens.13-15 By suppressing effector T-lymphocyte proliferation (i.e. CTL) an adequate tumor-specific immune response is prohibited by Treg cells and tumor growth is enabled. 16,1 7 Treg are present in increased numbers in peripheral blood and tumor tissue of patients with a variety of malignancies and associated with impaired prognosis.12• 1 6-21 When taking into account the importance of a careful balance between effector and suppressor responses in immune homeostasis, one might assume that an excess of Treg compared to CTL can facilitate the progressive growth of cancer. It has been shown in cervical and ovarian cancer that a high CTL/Treg ratio was significantly correlated with improved prognosis.7•22 Another lymphocyte subset of interest is memory T-lymphocytes. These T-cells arise after the primary response of the immune system against an antigen. In colorectal cancer, the presence of high levels of tumor infiltrating memory T-lymphocytes is correlated with a less advanced stage of disease, the absence of pathological evidence of early metastasis and increased survival.23

Limited data are available about Treg and CTL in endometrial cancer.10•24-26 Furthermore, to our knowledge, no study has been published yet on memory T-lymphocytes in endometrial cancer. The present study aimed at examining the presence and prognostic significance of these three TIL subsets in endometrial cancer.


Patients and methods

Patients From 1984 onwards tissue samples of patients with endometrial cancer treated at the Department of Gynecologic Oncology of the University Medical Center Groningen (UMCG) were collected and stored in the tissue storage system of the Pathology Department of the UMCG. For this study, we selected all patients from whom sufficient paraffin-embedded tumor tissue was available. Patients treated with radiotherapy prior to surgery or with a second malignancy were excluded. All patients underwent a total abdominal hysterectomy and a bilateral salpingo-oopherectomy (TAH-BSO). More extensive treatment with pelvic and/ or para-aortic lymph node dissection was performed in case of more advanced disease (stage II and higher) or unfavorable features (grade 3 tumor (including clear cell and serous papillary carcinomas)). Patients received adjuvant radiotherapy if indicated.27 Tumors were classified and graded by pathologists according to the World Health Organization (WHO) criteria.28 Staging occurred after surgical treatment according to the FIGO guidelines.29 Follow-up consisted of a clinical history, physical examination and a PAP smear of the vaginal vault at increasing intervals. Data were completed until September 2006.

Designing and constructing the TMA The tissue microarray (TMA) method allows the simultaneous evaluation of several markers on paraffin embedded tissues from hundreds of tumors.30 In brief, morphologically representative areas of tumor were marked on hematoxylin- and eosin- stained sections. Three core biopsies of 0.6 mm diameter each were randomly taken from the marked areas of tumor nests of the corresponding tissue block and placed in a recipient blank paraffin block (Shandon precision cut paraffin, No Bl002490, Thermo) on pre-defined array locations, using a precision instrument (Beecher Instruments, Silver Spring, Maryland). When all cores were inserted, recipient blocks were placed in a 37°C oven for fifteen minutes in order to maximize adhesion of the cores to the surrounding wax. Four µm thick sections were cut from the TMAs.

lmmunohistochemistry Mouse monoclonal antibodies recognizing Treg (anti-FoxP3 m22509; Abeam, Cambridge, UK), memory T-lymphocytes (anti-CD45R0 clone OPD4; Labvision, Fremont CA, USA) and CTL (anti-CDS clone C8/144B; DAKO, Beverlee, Belgium) were used for staining. In brief, sections were dewaxed in xylene and rehydrated through graded concentrations of ethanol. Antigen retrieval for Foxp3 was performed overnight at 80°C in TRIS-EDTA (pH 9.0). For CD45R0 and CDS, antigen retrieval was performed in citrate (pH 6.0) and TRIS/HCL (pH 9.0) respectively in the microwave for 15 minutes. After antigen retrieval, endogenous peroxidase activity was blocked in a 0.3% H202 solution for 30 minutes. All slides were incubated with the primary antibodies for 60 minutes at room temperature (dilutions: anti-FoxP3 1:100; anti CD45R0 1:50; anti-CDS 1:20). Sections incubated with anti-Foxp3 and anti-CD45R0 were subsequently incubated with a peroxidase-labelled polymer (Dako En Vision+ system). Sections with anti-CDS staining, RAMP0 (rabbit anti mouse peroxidase-labelled, DAKO, Beverlee, Belgium, 1:100)

25

Chapter 2

26

and GARP0 (goat anti-rabbit peroxidase-labelled, DAKO, 1:100) were used as secondary and tertiary antibodies respectively. Antigen-antibody reactions were visualized with NovaRED™ (Vector Laboratories, Burlingame, USA) for FoxP3 (Vector Laboratories, Burlingame) and 3,3' -diaminobenzidine was used for CDS and CD45R0 staining. Sections were counterstained with hematoxylin.

Analysis of immunohistochemistry staining Two independent observers (R.J. and N.L.) scored all immunohistochemistry stained slides without prior knowledge of clinicopathological information and counted the number of positively stained cells in tumor tissue (Figure 1). Staining was only analyzed when two or more cores were available, each containing more than 20% tumor tissue. In this way, the tissue samples resembled heterogeneity of whole tissue slides. Next, the average number of positively stained intra-tu.moral cells was calculated per 0.283 mm2 of tumor (i.e. one whole core consisting of 100% tumor tissue). In this way, we corrected for differences in sample size and we standardized the analysis.

Statistics Differences in clinicopathological parameters between groups were tested using the Chi square test. Kaplan-Meier analysis was used to estimate disease-free survival (DFS) and overall survival (OS) and survival differences between groups were assessed by Log-Rank test. DFS was defined as the date of diagnosis to date of recurrence. OS was defined as the date of diagnosis to the date of death due to endometrial cancer or date of last follow-up. Cox regression analysis was used for multivariate survival analyses. Variables that were significantly associated with survival in univariate analysis were entered into the multivariate analysis together with well-known prognostic factors. Covariates were eliminated from analysis using a manual backward method, until only independent predictors for survival remained. p<0.05 was considered statistically significant (tested 2-sided). All analyses were performed using SPSS software, version 14.0 (SPSS Inc., Chicago, IL, USA).

Figure 1. Immunohistochemical staining of tumor-infiltrating T-lymphocytes in endometrial cancer tissue. (A) Intra-tumoral CDB+ T-lymphocytes, (B) intra-tumoral FoxP3+ T-lymphocytes, (C) intra-tumoral CD45R0+ T-lymphocytes.


Results

Patients A total of 368 patients with endometrial cancer, treated between 1984 and 2004, was included in this study. Clinicopathological characteristics of these patients are summarized in Table 1. Median age at time of diagnosis was 65 years (range 32-89) and the median follow-up time was 4.4 years (range 0-21.5). The majority of patients was diagnosed

Table 1. Clinicopathological characteristics of 368 patients with endometrial cancer

Characteristics Patients no (n = 368) % *

Stage I 201 54.8 II 60 16.4 III 80 21.8 IV 26 7.0 Missing 1

Histologic grade 1 162 44.0 2 96 26.1 3 107 29.1 Undifferentiated 3 0.8

Histologic type Endometrioid 316 85.8 Serous papillary carcinoma 18 4.9 Clear cell carcinoma 26 7.1 Mixed tumours 5 1 .3 U ndifferen tia ted 3 0.8

Depth of myometrial invasion <112 210 58.0 >1/2 152 42.0 Missing 6

Lymphvascular space involvement Negative 245 71.6 Positive 97 28.4 Missing 26

Lymph node involvement Negative 114 71.7 Positive 45 28.3 Missing 209

Metastases Negative 306 83.8 Positive 59 16. 2 Missing 3

Recurrence No recurrence 286 77.7 Recurrence: 82 22.3

% Local 32 39.5 % Pelvic region 11 1 2.3 % Distant 39 48.1

* Percentages exclude missing cases

27

Chapter 2

28

with early stage disease (71.2%) and had well to moderately differentiated endometrial carcinomas (70.1 %). Type I endometrial cancer was diagnosed in 258 patients and type II in 110 patients. Hysterectomy and bilateral salpingo-oophorectomy was performed in 53.1% of the patients, more extended surgery with pelvic and/or para-aortic lymph node dissection was performed in 41.6% of the patients and 3.0% were treated otherwise. Adjuvant radiation therapy was indicated in 203 patients (55.2%). Recurrence of disease developed in 82 patients (22.3%), with a median time to recurrence of 1.51 years (range 0-7.6). Fifty-nine patients (16.0%) died as a result of endometrial cancer.

Association of TIL with clinicopathological parameters CDS+, FoxP3+ and CD45RQ+ T-lymphocytes were present in primary endometrial cancer tissue in 96.8%, 79.4% and 60.7% of patients, respectively. A clear binary distribution was present in CD45RQ+ T-lymphocytes, so these were categorized as 'absent' or 'present'. CDS+ and FoxP3+ cells however did not show such a clear distribution. A cut-off point was determined by calculating at which percentile the difference in survival between "high" and "low" group was the largest, as was previously done by Galon et al.5

For CDS+ cells, the 20th percentile (corresponding to 4 positively stained cells per 0.283 mm2 tumor tissue) was used as cut-off point: 'low numbers' was categorized as N<4 and 'high numbers' as N>4. For FoxP3+ T-lymphocytes the 80th percentile (corresponding to 8 positively stained cells per 0.283 mm2 tumor tissue) was used as cut-off point: 'low numbers' was defined as N<8 and 'high numbers' as N>8.

Associations of TIL with clinicopathological parameters are depicted in Table 2. High numbers of CDS+ T-lymphocytes were more frequently observed in early stage than in advanced stage disease (85.3% vs. 72.2%, p=0.006), as were low numbers of Treg (82.5% vs. 80.0%, p=0.022) and a high CDS+ /FoxP3+ ratio (56.5% vs. 33.0%, p<0.001). Compared to poorly differentiated tumors, well-differentiated tumors more often contained high numbers of CDS'" cells (85.3% vs. 72.4%, p=0.007), low numbers of FoxP3+ lymphocytes (86.4% vs. 68.4%, p=0.002), a high CDS+ /FoxP3+ ratio (56.0% vs. 34.4%, p<0.001) and presence of CD45RQ+ T-lymphocytes (65.4% vs. 50.5%, p=0.015). Furthermore, a high CDS+ /FoxP3+ ratio was associated with <50% myometrial invasion (55.6% vs. 42.3%, p=0.022), no lymphvascular space invasion (LVSI, 53.8% vs. 40.0%, p=0.034), negative lymph nodes (47.2% vs. 26.8%, p=0.028) and absence of distant metastases at time of diagnosis (52.9% vs. 35.1 %, p=0.016). Recurrence of disease occurred less often when tumors contained high numbers of CDs� T-lymphocytes (20.3% vs. 37.2%, p=0.006), a high CDS.,. /FoxP3+ ratio (16.1 % vs. 31.2%, p=0.002) or in the presence of CD45RO+ T-lymphocytes (17.1 % vs. 31.6%, p=0.004). Tumors of the endometrioid type were more frequently found in patients with high numbers of CDS+ T-lymphocytes (84.4% vs. 63.0%, p=0.001), a high CDS+ /FoxP3+ ratio (52.3% vs. 31.1 %, p=0.009) or when CD45RQ+ T-lymphocytes were present (66.4% vs. 28.9%, p<0.001).

1l 8-:� s � � ;::: ;::: ;::: ;::: (J} Q" ~ ~ · "1::l O O -

� � S 1 �- � � � � 4. :::-. · co :::r- nr � § s ::: ;r �

1 �1-

FIGO stage Stage 1/11 Stage III/IV

Differentiation grade Grade I/II Grade III/undifferentiated

Myometrial invasion < 50% > 50%

LVSI Absent Present

Lymph node involvement Negative Positive

Distant metastases Absent Present

Recurrence of disease No Yes

Tttmor type Endometrioid Non-endometrioid

CDS+ T-lymphocytes N=315

Low High

32 (14.7%) 186 (85 .3%) 27 (27.8%) 70 (72.2%)

32 (14.7%) 185 (85.3%) 27 (27.6%) 71 (72.4%)

28 (15.8%) 149 (84.2%) 29 (21 .6%) 105 (78.4%)

31 (15 .0%) 175 (85 .0%) 22 (24.7%) 67 (75.3%)

16 (17.4%) 76 (82.6%) 11 (25.6%) 32 (74.4%)

44 (17.2%) 212 (82.8%) 15 (26.3%) 42 (73.7%)

37 (15.4%) 204 (84.6%) 22 (29.7%) 52 (70.3%)

42 (15.6%) 227 (84.4%) 17 (37.0%) 29 (63.0%)

p value*

0.006

0.007

0.189

0.047

0.268

0.111

0.006

0.001

FoxPJ+ T-lympliocytes N=310

Low

179(82.5%) 66 (70.9%)

178 (83.6%) 67 (68.4%)

143 (81 .3%) 100 (76.9%)

165 (80.5%) 67 (77.0%)

72 (75.0%) 24 (63.2%)

205 (81.0%) 40 (71 .4%)

189 (78.8%) 56 (78.8%)

210 (79.5%) 35 (74.5%)

High

38 (17.5%) 27 (29.1 %)

35 (16.4%) 31 (31.6%)

33 (18.7%) 30 (23.1 %)

40 (19.5%) 20 (23.0%)

24 (25%) 17 (36.8%)

48 (19.0%) 16 (28.6%)

51 (21 .2%) 15 (21 .2%)

54 (20.5%) 12 (25.5%)

p value*

0.022

0.002

0.355

0.501

0 .054

0.109

0.982

0.433

FIGO = International Federation of GynaecologtJ and Obstetrics. LVSI = Lymphvascular space involvement * p-values were calculated using the Chi square-test. Bold signifies p<0.05 (tested 2-sided).

CDB+/FoxPJ+ ratio N=303

Low Hig}i p value*

91 (43.5%) 118(56.5%) <0.001 63 (67.0%) 31 (33.0%)

91 (44.0%) 116(56.0%) <0.001 63 (65 .6% 33 (34.4%)

75 (44.4%) 94 (55.6%) 0.022 75 (57.7%) 55 (42.3%)

92 (46.2%) 107 (53.8%) 0.034 51 (60%) 34 (40%)

47 (52.8%) 42 (47.2%) 0.028 30 (73.2%) 11 (26.8%)

115(47.1 %) 129(52.9%) 0.016 37 (64.9%) 20 (35.1 %)

106 (45.9%) 125(54.1 %) 0.002 48 (66.7%) 24 (33.3%)

123 (47.7%) 135 (52.3%) 0.009 31 (68.9%) 14 (31 .1 %)

CD45R0+ T-lymphocytes N=298

Absent Present

75 (36.4%) 131(63.6%) 42 (45.7%) 50 (54.3%)

71 (34.6%) 134 (65.4%) 46 (49.5% 47 (50.5%)

67 (40.6%) 98 (59 .4%) 47 (36.4%) 82 (63.6%)

73 (37.1 %) 124(62.9%) 33 (40.2%) 49 (59.8%)

39 (45 .3%) 47 (54.7%) 16 (37.2%) 27 (62.8%)

94 (38.8%) 148 (61.2%) 22 (40.7%) 32 (59.3%)

80 (34.8%) 150 (65 .2%) 37 (54.4%) 31(45.6%)

85 (33.6%) 168(66.4%) 32 (71 .1 %) 13 (28.9%)

p value*

0.131

0.015

0.466

0.617

0 .378

0.796

0.004

0.000

Chapter 2

30

Association of TIL with clinicopathological parameters in type J and type II tumors When analyzing patients with type I endometrial cancer, high numbers of FoxP3+ Tlymphocytes were more frequently present in advanced stage of disease (12.6% vs. 32.4, p=0.003). In the group of patients with type II endometrial cancer, high numbers of CDS+ T-lymphocytes were more often observed in early stage of disease (85% vs. 67.8%, p=0.021), and the presence of CD45R0+ cells was associated with less recurrence of disease (61.5% vs. 25%, p=0.001). Furthermore, a high CDS+ /FoxP3+ ratio was associated with early stage of disease (48.7% vs. 24.6%, p=0.014), <50% myometrial invasion (51.4% vs. 24.6%, p=0.008), negative lymph nodes (45.7% vs. 15.4%, p=0.013) and less recurrence of disease (41.2% vs. 17.9%, p=0.029).

Survival analysis Endometrial cancer recurred in 82 patients (22.3%), with local/vaginal recurrence in 39.5 %, pelvic recurrence in 12.3% and distant recurrence in 48.1 % of these patients. Time to recurrence was longer when high numbers of CDS+ T-lymphocytes were present (mean 2.3 vs. 1.4 years, p=0.019, Figure 2A). Similarly, a high CDS+ /FoxP3+ ratio was associated with prolonged DPS (mean 2.8 vs. 1.7 years, p=0.007, Figure 2B). Similar to the time to recurrence, high numbers of CDS+ T-lymphocytes were associated with a better OS (mean 17.2 vs. 9.8 years, p<0.001, Figure 2C). Furthermore, patients with intratumoral CD45R0+ T-lymphocytes had longer OS (mean 17.3 vs. 14.6 years, p=0.023, Figure 2D). The presence of high numbers of FoxP3+ lymphocytes was associated with decreased OS (mean OS 14.1 vs. 18.2 years, p=0.034) in the subgroup of patients with a type I tumor. Furthermore, analogous to the whole cohort, type I patients with a high CDS+/ FoxP3+ ratio had a longer recurrence free interval (mean DPS 3.3 vs. 2.0 years, p=0.011). Overall survival of type II endometrial cancer patients was positively associated with high numbers of CDS+ lymphocytes (mean OS 15.9 vs. 4.1 years, p<0.001) and the presence of CD45R0+ T-lymphocytes (mean 11.1 vs. 9.2 years, p=0.006).

Multivariate analysis Next to tumor-infiltrating lymphocytes, well-known prognostic factors, such as age, FIGO stage, differentiation grade, histological type and myometrial invasion were entered into a multivariate Cox regression analysis. The presence of high numbers of intra-tumoral CDS+ T-lymphocytes was an independent predictor of increased OS after adjustment for the above-mentioned well-known prognostic factors (HR 0.48, 95% C.I. 0.26-0.89, p=0.019). Furthermore, the presence of a high COS+ /FoxP3+ ratio was an independent predictor for increased DPS in type I endometrial cancer patients (HR 0.44, 95% C.I. 0.23-0.84, p=0.013). In type II patients, CO45RO+ T-lymphocytes, age at diagnosis and stage of disease were independent predictors of OS (HR 0.42, 95% C.I. 0.19-0.93, p=0.033). Interestingly, in the model incorporating COS+ T-lymphocytes and adjusting for some well-known prognostic factors, these cells were the only independent predictor influencing OS in type II endometrial cancer patients (HR 0.17, 95% C.I. 0.08-0.36, p<0.001).


..

..

op P-0.019

A o.o

,p

.. : ! •11 . i r "

op P<0.001

C •.a

cos+ T -lymphocytes

High numbers -- Low numbers

o.a P-0.001

1,0 :.o lA 4.G

disease freo Interval (y .. rs)

u B OP

cos+ T -lymphocytes

--

◄.O ,.o follow-up (yurs)

High numbers Low numbers

... ,op

,p

0,11

;;

! J: D.t . i :i O,◄

!i u 02

op P.O.D2J

D u

Discussion

COs+/FoxP3• ratio

High ratio

1.0 2/J 3.0 •.D

disease frH Interval (y■ars)

...... ___

SP

CD45Ro+ T-lymphocytes

�p 4!1.D

follow-up (years)

--

..

Present Absert

10P

Figure 2. Kaplan-Meier survival curves with Log-Rank tests in the entire cohort of endometrial cancer patients.

We studied three different subsets of TIL in a large and well-documented population of endometrial cancer patients. The presence of high numbers of COS+ T-lymphocytes was an independent predictor of improved overall survival in the entire cohort as well as for the subgroups of type II endometrial cancer patients. A high COS+ /FoxP3+ ratio was independently associated with improved survival in type I endometrial cancer patients. Overall survival in type II patients was further independently predicted by the presence of C045RO+ cells.

To our knowledge, this is the first study to examine the prognostic influence of tumor-infiltrating COS+ cytotoxic T-lymphocytes, FoxP3+ regulatory T-lymphocytes and C045RO+ memory T-lymphocytes in such a large, unselected and representative study population of endometrial cancer patients. We show that high numbers of intra-tumoral COS+ T-lymphocytes have a positive prognostic influence in endometrial cancer patients in general, as well as in a subgroup of patients with type II endometrial cancer. The former result has been shown in two previous studies.10,26 The number of patients included however was much lower compared to our study population and not all FIGO stage or tumor types were represented. Furthermore, both studies selected three areas of tumor-tissue with the most abundant TIL which might not be representative for the entire tumor.

31

Chapter 2

32

Surprisingly, high numbers of CDB+ cytotoxic T-lymphocytes were more often found in low grade and early stage endometrial cancer patients. Whether the difference in infiltration of T-cells in low grade versus high grade tumors is due to factors produced by the cancer cells themselves or is correlated to MHC expression31 needs to be elucidated. Where other studies show a clear association between the presence of Treg and impaired survival in cancer patients12•16-21 the presence of FoxP3+ T-lymphocytes in our study population was only associated with decreased overall survival in type I endometrial cancer patients. As expected, the presence of high numbers of FoxP3+ T-lymphocytes was associated with unfavorable clinicopathological factors in endometrial cancer patients. We used FoxP3 as marker for Treg since it is currently considered to be the most specific marker for Treg, as expression is almost exclusively restricted to the CD4+CD25+ Treg.32 The assumption that Treg have a negative influence on survival has been confirmed in several types of cancer16-19•21, and was recently also described in stage I endometrial cancer patients.25

As was first described in ovarian cancer7, the balance between effector and suppressor immune cells is important. We show that a high CD8+ /FoxP3+ ratio is important in endometrial cancer as well, where a high ratio is associated with all favorable parameters tested. More importantly, a high CD8+ /FoxP3+ ratio is the only independent predictor of increased disease free survival in type I endometrial cancer patients in this study.

Previous studies have shown a positive influence of tumor-infiltrating memory T-lymphocytes on survival of colorectal cancer patients.5•23 Although CD45R0 is not exclusively expressed on memory T-lymphocytes where expression also occurs on activated T-cells, it is currently one of the most specific markers for memory T-lymphocytes.5•23

For the first time, we show that in endometrial cancer, intra-tumoral CD45RQ+ T-lymphocytes are an independent predictor of increased overall survival in type II endometrial cancer. Interestingly, recurrence of disease less often occurred in the presence of CD45RQ+ T-lymphocytes. This observation is in line with the functional mechanism of memory T-lymphocytes, namely providing a more rapid and effective secondary immune response to previously encountered antigens.

Over the years, progress has been made in optimizing diagnosis and treatment of endometrial cancer, which has resulted in a generally good prognosis in early-diagnosed disease.1 Patients with unfavorable prognostic factors (high stage of disease, deep myometrial invasion and high grade tumors) are currently treated with more advanced surgery and/ or adjuvant radiotherapy. Still a significant number of women die of this disease each year. Further optimized treatment is therefore warranted, especially for the group of endometrial cancer patients with unfavorable prognostic factors. Therefore, immunotherapeutic strategies need to be further explored in endometrial cancer. Targeting FoxP3 will offer a strategy to control Treg over time. 33 Furthermore, strategies inducing tumor-specific CTL need to be explored.26•34

In conclusion, we show that the presence of high numbers of CDS+ T-lymphocytes are an independent predictor of improved OS for endometrial cancer patients in general as well as for type II endometrial cancer patients. A high CD8+ /FoxP3+ ratio is an


independent prognostic factor for a better DFS. Patients with intratumoral CD45R0+

T-lymphocytes less often develop recurrence disease and have longer OS. Our data support the general idea that the immune system has an important role in endometrial cancer. These findings suggest that the immunochistochemical evaluation of immune cells should be further explored in the development for novel therapeutic strategy for endometrial cancer.

Acknowledgements

This study was partly funded by Dutch Cancer Society grants 2002-2768 (H.W. Nijman) and 2007-3919 (H.W. Nijman/ N.Leffers).

33

Chapter 2

34

References

1 . Amant F, Moerman P, Neven P, Timmerman D, Van Lirnbergen E, Vergote I . Endometrial can-cer. Lancet 2005;366:491-505.

2. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2008. CA Cancer J Clin 2008;58:71-96.

3. Bokhman JV. Two pathogenetic types of endometrial carcinoma. Gynecol Oncol 1983;15:10-7.

4. Clemente CG, Mihm MC, Jr., Bufalino R, Zurrida S, Collini P, Cascinelli N. Prognostic value of tumor infiltrating lymphocytes in the vertical growth phase of primary cutaneous melanoma. Cancer 1996;77(7):1303-10.

5. Galon J, Castes A, Sanchez-Caba F, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science 2006;313:1960-4.

6. Halpern AC, Schuchter LM. Prognostic models in melanoma. Semin Oncol 1997;24(1 Suppl 4):S2-S7.

7. Sato E, Olson SH, Ahn J, et al. Intraepithelial CDS+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proc Natl Acad Sci U S A 2005;102:18538-43.

8. Vesalainen S, Lipponen P, Talja M, Syrjanen K. Histological grade, perineural infiltration, tumour-infiltrating lymphocytes and apoptosis as determinants of long-term prognosis in prostatic adenocarcinoma. Eur J Cancer 1994;30A:1797-803.

9. Zhang L, Conejo-Garcia JR, Katsaros D, et al. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med 2003;348:203-13.

10. Kondratiev S, Sabo E, Yakirevich E, Lavie 0, Resnick MB. Intratumoral CDS+ T lymphocytes as a prognostic factor of survival in endometrial carcinoma. Clin Cancer Res 2004;10:4450-6.

11. Naito Y, Saito K, Shiiba K, et al. CDS+ T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer. Cancer Res 1998;58:3491-4.

12. Wolf AM, Wolf D, Steurer M, Gastl G, Gunsilius E, Grubeck-Loebenstein B. Increase of regulatory T cells in the peripheral blood of cancer patients. Clin Cancer Res 2003;9:606-12.

13. Sakaguchi S. Nahtrally arising CD4+ regulatory t cells for immunologic self-tolerance and negative control of immune responses. Annu Rev Immunol 2004;22:531-62.

14. Fehervari Z, Sakaguchi S. CD4+ Tregs and immune control. J Clin Invest 2004;114:1209-17.

15. Yamaguchi T, Sakaguchi S. Regulatory T cells in immune surveillance and treatment of cancer. Semin Cancer Biol 2006;16:115-23.

16. Curiel TJ, Coukos G, Zou L, et al. Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nat Med 2004;10:942-9.

17. Kono K, Kawaida H, Takahashi A, et al. CD4(+)CD25high regulatory T cells increase with rumor stage in patients with gastric and esophageal cancers. Cancer Immunol Immunother 2006;55:1064-71 .

18. Schaefer C, Kim GG, Albers A, Hoermann K, Myers EN, Whiteside TL. Characteristics of CD4+CD25+ regulatory T cells in the peripheral circulation of patients with head and neck cancer. Br J Cancer 2005;92:913-20.

19. Bates GJ, Fox SB, Han C, et al. Quantification of regulatory T cells enables the identification of high-risk breast cancer patients and those at risk of late relapse. J Clin Oncol 2006;24:5373-80.

20. Woo EY, Chu CS, Goletz TJ, et al. Regulatory CD4(+)CD25(+) T cells in tumors from patients with early-stage non-small cell lung cancer and late-stage ovarian cancer. Cancer Res 2001;61 :4766-72.

Presence of tumor-infiltrating lymphocytes is an independent prognostic factor in hJpe I and II endometrial cancer

21. Wolf D, Wolf AM, Rurnpold H, et al. The expression of the regulatory T cell-specific fork.head box transcription factor FoxP3 is associated with poor prognosis in ovarian cancer. Clin Cancer Res 2005;11:8326-31.

22. Piersma SJ, Jordanova ES, van Poelgeest MI, et al. High number of intraepithelial CDS+ tumorinfiltrating lymphocytes is associated with the absence of lymph node metastases in patients with large early-stage cervical cancer. Cancer Res 2007;67:354-61 .

23. Pages F, Berger A, Camus M, et al . Effector memory T cells, early metastasis, and survival in colorectal cancer. N Engl J Med 2005;353:2654-66.

24. Fattorossi A, Battaglia A, Ferrandina G, et al . Lymphocyte composition of tumor draining lymph nodes from cervical and endometrial cancer patients. Gynecol Oncol 2004;92:106-15.

25. GiatromanolakiA, Bates GJ, Koukourakis MI, et al. The presence of tumor-infiltrating FOXP3+ lymphocytes correlates with intratumoral angiogenesis in endometrial cancer. Gynecol Oncol 2008;110:216-21.

26. Ino K, Yamamoto E, Shibata K, et al. Inverse Correlation between Tumoral Indoleamine 2,3-Dioxygenase Expression and Tumor-Infiltrating Lymphocytes in Endometrial Cancer: Its Association with Disease Progression and Survival. Clin Cancer Res 2008;14:2310-7.

27. Scholten AN, van Putten WL, Beerman H, et al . Postoperative radiotherapy for Stage 1 endometrial carcinoma: long-term outcome of the randomized PORTEC trial with central pathology review. Int J Radiat Oncol Biol Phys 2005;63:834-8.

28. Tavassoli FA, Devilee P. World Health Organization: Tumours of the Breast and Female Genital Organs (WHO Classification of Tumours). 1 ed. IARC Press-WHO; 2003.

29. Cancer Comittee of the International Federation of Gynaecology and Obstetrics (1986). Gynecol Oncol 2008;25:383-5.

30. Kononen J, Bubendorf L, Kallioniemi A, et al. Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nat Med 1998;4:844-7.

31. Han LY, Fletcher MS, Urbauer DL, et al. HLA class I antigen processing machinery component expression and intratumoral T-Cell infiltrate as independent prognostic markers in ovarian carcinoma. Clin Cancer Res 2008;14:3372-9.

32. Hori S, Nomura T, Sakaguchi S. Control of regulatory T cell development by the transcription factor Foxp3. Science 2003;299:1057-61.

33. Nair S, Boczkowski D, Fassnacht M, Pisetsky D, Gilboa E. Vaccination against the fork.head family transcription factor Foxp3 enhances tumor immunity. Cancer Res 2007;67:371-80.

34. Santin AD, Bellone S, Ravaggi A, Roman JJ, Pecorelli S, Parham GP, et al. Induction of tumourspecific CDS(+) cytotoxic T lymphocytes by tumour lysate-pulsed autologous dendritic cells in patients with uterine serous papillary cancer. Br J Cancer 2002;86:151-7.

35

C_hapter 3

Loss of HLA class I and mismatch repair protein

expression in sporadic endometrioid endometrial

carcinomas

Renske A. de Jong

Annemarie Boerma

H. Marike Boezen

Marian J.E. Maurits

Harry Hollema

Hans W. Nijman

International Journal of Cancer

2012;22:1177-86.

Loss of HLA class I and mismatch repair protein expression in sporadic endometrioid endometrial carcinomas

Abstract

Tumor cells can escape from cytotoxic T-cell responses by downregulation of human leukocyte antigen (HLA) class I molecules expressed at the cell surface which has been associated with a deficient mismatch repair (MMR) system in colorectal carcinomas. This study investigated the association between expression of MMR proteins and HLA class I in sporadic endometrioid endometrial carcinomas (EC). In a consecutively selected cohort of 486 EC patients, MMR proteins (MLHl, MSH2 and MSH6) and HLA class I (HLA-A, -B, -C or �

2m) were investigated by immunohistochemistry. Expression

levels of MMR proteins and HLA class I were compared between low-grade and highgrade EC's. HLA class I expression was compared between tumors with loss (negative immunostaining of ;;:,:1 MMR protein) and expression of MMR proteins. Associations between previously determined numbers of intratumoral CDS+ T-lymphocytes and expression of MMR proteins and HLA class I and the influence on survival was determined. EC's with loss of MMR protein expression (33.5%) more frequently have loss of HLA-B/C (37.3%), compared to EC's with MMR protein expression (25.5%, p=0.007). Patients with loss of MMR proteins have a worse disease-specific survival compared to patients with expression (p=0.039). CDS+ T-lymphocytes have a positive influence on disease-free and disease-specific survival in the total EC cohort but not in patients with loss of MMR protein expression. In conclusion, our results indicate that loss of MMR protein expression is related to selective downregulation of HLA class I which contributes to immune escape in EC with an abnormal MMR system.

39

Chapter 3

40

Introduction

Endometrial carcinoma (EC) is the most common gynecological malignancy and the fourth most common malignancy amongst women in the developed world after breast, lung and colorectal cancer.1 Based on clinicopathological, immunohistochemical and molecular genetic studies, EC can be divided in two types; type I consists of endometrioid carcinomas and type II consist of clear-cell and serous papillary carcinomas.2•3 In type I EC, microsatellite instability (MSI) is a frequent phenomenon with incidences ranging from 20-30% compared to 0-11 % in type II EC.4•5 MSI is caused by a failure of the mismatch repair (MMR) system to cut out and replace the mismatching DNA strains. MMR deficiency leads to accumulation of single base-pair mismatches, as well as small insertions and deletions in tandem repeats known as microsatellites.5

In colorectal cancer, a MSI phenotype is associated with improved prognosis.6 Possibly, the mutation phenotype not only affects oncogenes but also increases the production of abnormal peptides able to elicit specific cytotoxic immune responses against tumor cells.7•8 Colorectal carcinomas with a MSI+ phenotype are characterized by increased numbers of tumor-infiltrating lymphocytes (TIL) 9•1 0 which are activated cytotoxic Tlymphocytes (CTL) in the majority of the cases.7 MSI status has not been associated with improved survival in EC and results on associations between MSI and clinicopathological characteristics such as FIGO stage, tumor grade and myometrial invasion differ between studies (reviewed in 5•11) .

The mainstay of a tumor-specific immune response is through the human leukocyte antigen (HLA) class I system present on the cell surface of every human cell. HLA class I comprises the classical (class Ia) HLA-A, -B and -C antigens and the nonclassical (class lb) HLA-E, -F and -G proteins. A fully assembled HLA class I molecule is a heterotrimer consisting of a heavy a chain, a nonpolymorphic �2-microglobulin (�2m) light chain and a bound antigenic peptide. Tumor cells present antigen-derived peptides which are, when conjugated with HLA class I molecules, specifically recognized by CTLs which can directly kill the tumor cell after recognition.12 However, tumor cells develop several mechanisms to escape immune responses, for example by modifying HLA class I expression on their cell surface. In this way, tumor cells are protected from specific recognition and killing by immune cells. Selective loss of the HLA haplotype, specific loci or alleles are different mechanisms which can cause aberrant expression of HLA class I antigens.8•12•13

As shown in colorectal cancer, loss of HLA class I expression occurs more frequently in MSI+ tumors compared to microsatellite stable (MSS) tumors.8•1 4 In this way, MSI+

colorectal carcinomas develop mechanisms to escape immune responses by a selective pressure.7•8 This study investigated the occurrence of loss of MMR proteins (negative immunostaining of MLHl, MSH2 or MSH6) and HLA class I expression in a consecutively selected cohort of sporadic endometrioid EC. Since it is known that high-grade EC (tumor grade 3) behave more aggressively than low-grade EC (tumor grade 1 or 2), a distinction was made between low-grade and high-grade EC's. Frequencies of HLA class I downregulation were compared between EC without MMR protein expression


Table 1. Clinicopathological characteristics of patients with EC All patients (%)1 (n = 486)

Age, median (IQR), years 66 (57-75)

FICO stage I 339 (69.8%) II 63 (13.0%) III 71 (14.6%) IV 13 (2.6%) Missing 0

Dunor grade Grade 1 241 (49.6%) Grade 2 163 (33.5%) Grade 3 82 (16.9%) Missing 0

Myometrial invasion <50% 275 (56.6%) >50% 211 (43.4%) Missing 0

LVSI Negative 373 (79.2%) Positive 98 (20.8%) Missing 15

Lymph node status Negative 100 (78.1 %) Positive 28 (21 .9%) Missing 358

Treatment TAH + BSO 340 (70.0%) TAH + BSO + LND 128 (26.3%) Other 18 (3.7%) Missing 0

Adjuvant radiotlterapy No 217 (44.7%) Yes 269 (55.3%) Missing 0

Recurrent disease No 390 (80.6%) Yes 94 (19.4%) Loco regional 40 (42.6%) Distant 54 (57.4%) Missing 2

Follow-up status Alive 363 (74.8%) Death due to disease 53 (10.9%) Death due to other disease 69 (14.3%) Missing 1

1Percentages exclude missing cases. LVSI: Lymphvascular space involvement. TAH + BSO: total abdominal hysterectomy with bilateral salpingo-oopherectomy. LND: lymph node dissection

41

Chapter 3

42

and those with expression. Numbers of intratumoral CDS+ T-lymphocytes were related to expression status of MMR proteins and HLA class I and the influence on survival was determined.

Material and methods

Patients For our study, tissue material and clinicopathological characteristics (see Table 1) of EC patients were collected as previously reported by our group.15,16 Patients diagnosed with a non-endometrioid tumor type (n=57) or patients with a genetically proven Lynch syndrome-related EC (n=ll) were excluded. Ultimately, 486 patients with sporadic endometrioid EC (further referred to as: EC) were included, all had been treated with primary surgery in the University Medical Center Groningen (UMCG) or the Medical Center Leeuwarden (MCL) (a large teaching hospital). Adjuvant radiotherapy was given if indicated.17 Patients were followed up to 10 years with gradually increasing intervals and follow-up data were updated until August 2010.

Institutional review board approval All relevant data were retrieved from our computerized database and transferred into a separate, anonymized, password protected database. Patient identity was protected by study-specific, unique patient codes, which were only known to two dedicated data managers, who also have daily responsibility for the larger database. In case of uncertainties with respect to clinicopathological and follow-up data, the larger databases could only be checked through the data managers, thereby ascertaining the protection of patients' identity. Due to these procedures according to Dutch law no further patient or institutional review board approval was needed (http:/ /www.federa.org/).

Antibodies For detection of a wide range of HLA class I heavy chains with a few antibodies, the mouse monoclonal HCA2, HClO, which recognize HLA class I heavy chains (kindly provided by Prof. Dr. J. Neefjes), were used. The reactivity spectrum of HCA2 is composed of all HLA-A chains (except HLA-A24), as well as some HLA-B, HLA-C, HLAE, HLA-F, and HLA-G chains.18,19 HClO reacts mostly with HLA-B and HLA-C heavy chains and some HLA-A (HLA-AlO, HLA-A28, HLA-A29, HLA-A30, HLA-A31, HLAA32, HLA-A33.20,21 In order to estimate the expression of �

2m light chain we used the

rabbit polyclonal anti-�2m antibody, purchased from DAKO (Copenhagen, Denmark).

For recognition of cytotoxic T-lymphocytes, mouse monoclonal antibodys (anti-CDS clone C8/ 144B; DAKO, Heverlee, Belgium) were used.

Immunohistochemistry For staining, 4 mm sections were cut from previously constructed tissue microarrays (TMA) of formalin-fixed paraffin-embedded tumors15,16 and applied to 3-amino-propyltriethoxy-silane coated glass slides (Sigma-Aldrich, Diesenhofen, Germany). Antibo-


Figure 1. Representative immunohistochemistry staining of HCA2 (A-C), HLA-B/C (D-F) and l32m (G-1). Loss of expression (A,D,G), partial loss of expres-sion (B,E,H) and normal expression (C,F,I).

dies for immunostaining of CDS+ T-lymphocytes, HLA-B/C and f\m were applied and reported previously.15-16 Immunostaining for HLA-A was performed by dewaxing and rehydrating TMA sections. Microwave antigen retrieval was performed in citrate (pH 6.0). Endogenous peroxidase activity was blocked with 0.3% Hp

2 for 30 minutes after

which slides were incubated with the primary HCA2 antibody (dilution: 1:100) for 60 minutes. Sections were incubated with RAMpo (rabbit anti-mouse peroxidase-labeled, DAKO, Beverlee, Belgium, 1:100) and GARP0 (goat anti-rabbit peroxidase-labeled, DAKO, Beverlee, Belgium, 1:100). Antigen-antibody reactions were visualized with 3,3-diaminobenzidine, and sections were counterstained with haematoxylin. Monoclonal antibodies were used against MLHl (prediluted) (clone G168-728, Ventana), MSH2 (1:400) (clone G219-1129, Ventana) and MSH6 (1:400) (clone G70220-15049, Transduction laboratories). Staining was performed automatically with Ventana BenchMark®

ULTRA IHC/ISH Staining Module according to the manufacturer 's instructions.

Immunohistochemistry stained slides were scored independently by two investigators (RJ and HH) and was only analyzed when two or more tissue microarray cores were available, containing at least 20% tumor tissue. MLHl, MSH2 and MSH6 expression was scored as either negative (i.e. total absence of detectable nuclear staining of tumor cells) or positive. HCA2 expression was determined using a semiquantitative scoring system where both intensity and percentage of positive tumor cells were analyzed. The intensity of staining was scored as O for absent, 1 for weak, 2 for positive or 3 for strong positive expression, respectively. The percentage of positive tumor cells was scored

43

Chapter 3

44

as O for 0%; 1 for �1-5%; 2 for �5-25%; 3 for �25-50%; 4 for �50-75% and 5 for �75-100% (Figs. lA-lC). The sum of both scores was used to identify three categories of expression: normal expression (total score 6.5-8), partial loss (2.5-6.5) and total loss (0-2.5). Immunostaining for CDS+ T-lymphocytes, HLA-B/C and i\m was scored and reported previously by our group (Figs. lD-lF and figs. lG-lI, respectively).15•16

Statistical analysis All continuous variables were checked for normality of the distribution using P-P plots. In case of skewed distributions, the median and interquartile ranges (IQR, 25th to 75th

percentile) were presented. Staining of HLA class I antigens was dichotomized for further analysis; partial loss and normal expression were taken together and compared to total loss of expression.14 Chi-square tests (or Fisher's exact tests, if appropriate) were used to test for differences between expression of MMR proteins and HLA class I and for differences of expression levels of MMR proteins and HLA class I between lowgrade and high-grade EC. Mann-Whitney U tests were used to compare numbers of CDS+ T-lymphocytes between tumors with loss (negative immunostaining of MLHl, MSH2 or MSH6) and expression of MMR proteins (positive immunostaining of MLHl, MSH2 and MSH6) as well as between tumors with loss or presence of HLA class I expression. Associations between clinicopathological characteristics (see Table 1) and MMR protein expression were estimated using univariate logistic regression analysis. Immunohistochemistry expression was used as dependent and clinicopathological characteristics were used as independent variables; odds ratios (ORs) and 95%-confidence intervals (95%-CI) were estimated using logistic regression analysis. Differences in disease specific survival (DSS) and disease-free survival (DFS) were plotted using Kaplan-Meier survival curves and evaluated by log-rank tests. DSS was defined as date of diagnosis until death due to EC or the date of last follow-up and DFS was defined as date of diagnosis until recurrent disease, metastasis, death due to EC or last date of follow-up. All tests were performed two-sided and p-values of <0.05 were considered statistically significant. Analyses were performed using the software package PASW, version 18.0 for Windows (SPSS Inc. Chicago Illinois, USA).

Results

Patients Patients were diagnosed with early stage of disease (FIGO I/II) in the majority of the cases (82.8%) and a low tumor grade (grade 1 and 2) (83.1 %). Median age at time of diagnosis was 66 years (IQR: 57-75). Recurrent disease was diagnosed in 94 of 497 patients (19.4%) with a median time until recurrent disease of 25.4 months (IQR: 12.7-42.9). Median time of follow-up was 5.7 years (IQR: 2.8-8.8) during which 53 of 486 patients (10.9%) died as a result of EC.


MMR protein expression Negative MLHl expression was observed in 23.8%, negative MSH2 expression in 14.8% and negative MSH6 expression in 9.3% of the patients (Table 2). Loss of MMR protein expression was observed in 33.5%, concurrent negative expression for 2 MMR proteins occurred in 11.3% and triple negative MMR protein expression was observed in 2.6% of the patients. In the majority of the cases, expression patterns were similar: for MSH2 and MSH 6 in 90.0% (p<0.001), for MLHl and MSH2 in 73.2% (p<0.001) and for MLHl and MSH6 in 72.3% (p=0.028) of the cases (data not shown). Expression levels of MMR

Table 2. Immunohistochemistry expression of MMR proteins in EC patients.

All cases Low-grade High-grade p (n=486) (n=404) (n=82) value1

0.962 MLH1 negative 108 (23.8%) 89 (23.8%) 19 (24.1 %)

positive 345 (76.2%) 285 (76.2%) 60 (75.9%) missing 33 30 3

0.559

MSH2 negative 68 (14.8%) 58 (15.2%) 10 (12.7%) positive 392 (85.2%) 323 (84.8%) 69 (87.3%) missing 26 23 3

0.881 MSH6 negative 43 (9.3%) 36 (9.4%) 7 (8.9%)

positive 419 (90.7%) 347 (90.6%) 72 (91 .1 %) missing 24 21 3

0.885 MMR (l) negative 155 (33.5%) 128 (33.3%) 27 (34.1 %)


0.460 MMR (2) negative 52 (11.3%) 45 (11.7%) 7 (8.9%)

positive 410 (88.7%) 338 (88.3%) 72 (91.1 %) missing 24 21 3

1.000 MMR (3) negative 12 (2.6%) 10 (2.6%) 2 (2.5%)


1 Comparison between low-grade and high-grade EC. Low-grade: tumor grade 1 or 2 High-grade: tumor grade 3 MMR (1): concurrent negative expression of at least 1 MMR protein MMR (2): concurrent negative expression of at least 2 MMR proteins MMR (3): concurrent negative expression of 3 MMR proteins Missing: cores without sufficient qualihJ (�2 cores containing more than 20% tumor tissue) for evaluation of immunohistochemistn; expression.

45

Chapter 3

46

! •

- lfflC#�� - �et� ...........

- �-•-�'1.41•CN\liq'\ - "'�·""""""-""'kn

A

r r r ! J . J .

- c:,a., 1�.-.-- (ta.. t�--- - � r�fl!O(�-

, ..... , -�,�--- . , ..... 7 - cN-t�..,. . , .. ..,, I - �f�M(tt,4•

c · .

D __ E -... -

r �

A .

_ ,,,.. , __ , - C:.• i...U•,,...- - c:t:4• '�-"'"-. , ..... J --·-... , . .. - J - cm• r�• , ..... , -a::.-'� .... . .. F ...... _ G -- H ........ ,,. ... )

Figure 2. Influence of MMR protein expression on survival in EC patients; DSS (A) and DFS (B). Influence of the number of intra-tumoral COB+ T-lymphocytes on survival; the entire cohort of EC patients (DSS: C, DFS: F), EC with MMR protein expression (DSS: D, DFS: G) and EC with negative MMR protein expression (DSS: E, DFS: H)

proteins were not significantly different between low-grade and high-grade EC. Negative MSH6 expression was associated with age (OR: 0.97; 95%-CI: 0.95-0.99, p=0.037) and advanced stage of disease (OR: 1.98; 95%-CI: 1.01-3.86). No association was found between treatment details, clinicopathological characteristics and loss of MMR protein expression or more specifically, negative MLHl or MSH2 expression (data not shown). Patients with loss of MMR protein expression had a worse DSS compared to patients with expression of MMR proteins (p=0.039) (Figure 2A) in contrast to DFS (p=0.148) (Figure 2B).


HLA class I expression Loss of HLA class I heavy chain (HLA-A and/or HLA-B/C) was observed in 41.3% of the patients (Table 3). Selective loss of HLA-A or HLA-B/C occurred in 31.7% and 29.2%, respectively. Loss of �2

m expression was observed in 1.1 % and was always accompanied with loss of HLA class I heavy chain. Patients with loss of HLA-B/C expression more often had received adjuvant radiotherapy (p=0.007). Loss of HLA class I expression was more frequently observed in high-grade tumors compared to lowgrade tumors (p=0.026) just as selective loss of HLA-B/C (p<0.001). HLA-A and �

2m

expression did not differ between high-grade and low-grade EC.

Table 3. HLA class I downregulation in EC patients

All cases (n=486)

HLA class I, loss 195 (41.3%) heavy chain expression 277 (58.7%)

missing 14

HLA-A loss 139 (31.7%) expression 299 (68.3%) missing 48

HLA-B/C negative 134 (29.2%) positive 325 (70.8%) missing 27

�2m negative 5 (1.1 %) positive 441 (98.9%) missing 40

1 Comparison between low-grade and high-grade EC Low-grade: tumor grade 1 or 2 High-grade: tumor grade 3 Bold values indicate p<0.05

Low-grade High-grade p (n=404) (n=82) value1

0.026 153 (39.0%) 42 (52.5%) 239 (61.0%) 38 (47.5%) 12 2

0.159 109 (30.3%) 30 (38.5%) 251 (69.7%) 48 (61 .5%) 44 4

<0.001 97 (25.5%) 37 (47.4%) 284 (74.5%) 41 (52.6%) 23 4

0.194 3 (0.8%) 2 (2.7%) 369 (99 .2%) 72 (97.3%) 32 8

Missing: cores without sufficient quality (<!:2 cores containing more than 20% tumor tissue) for evaluation of immunohistochemistn; expression.

47

Chapter 3

48

Association between MMR and HLA class I expression We analyzed whether HLA class I expression differed between tumors with loss or expression of MMR proteins (see Table 4). Tumors with negative MSH2 expression more often showed loss of HLA class I heavy chain (52.9%) compared to MSH2 positive tumors (39.5%) (p=0.039). Tumors with loss of 2".l MMR protein tended to have lost HLA class I expression more frequently (p=0.070) whereas concurrent negative expression of 2 MMR proteins was significantly associated with loss of HLA class I expression (p=0.011). Loss of HLA-B/C expression was more frequently observed in tumors with negative staining for MLHl (39.3%) compared to MLHl positive staining (26.4%) (p=0.011). Furthermore, loss of HLA-B/C was associated with loss of at least one or two MMR proteins (p=0.007 and p=0.014, respectively). HLA-A or �2m expression did not significantly differ between tumors with loss and expression of MMR proteins.

Association between CDS+ T-lymphocytes, HLA class I and MMR protein expression The number of tumor-infiltrating CDS+ T-lymphocytes was determined in a previous study and represents a subgroup of the current study cohort (n=253).16 The median number of intratumoral CDS+ T-lymphocytes per tissue core was 16 (IQR: 6-35). The number of CDS+ T-lymphocytes was significantly lower in tumors with loss of HLA class I heavy chain (median: 11; IQR: 3-27) compared to tumors with expression (median: 19; IQR: 9-40) (p<0.001). Furthermore, numbers of CDS+ T-lymphocytes were significantly lower in tumors with selective loss of HLA-A (median: 12; IQR: 3-27) or HLA-B/C (median: 10; IQR: 3-30) compared to tumors with expression of these antigens (HLA-A: median: 17; IQR: 8-36) and (HLA-B/C: median: 17; IQR: 8-37) (p=0.015 and p=0.005, respectively). The number of CDS+ T-lymphocytes did not differ between cases with loss and expression of �2m or between tumors with loss or expression of MMR proteins (data not shown).

Association between tumor-infiltrating CDS+ T-lymphocytes and survival In the total cohort, a high number of CDS+ T-lymphocytes was associated with better DSS (p<0.001) and longer DFS (p<0.001) (Figs. 2C and 2F, respectively) compared to low numbers of CDS+ T-lymphocytes. This significant association between high numbers of CDS+ T-lymphocytes and better DSS and longer DFS was also found in a subgroup of these patients with MMR protein expression (n=165) (both: p<0.001) (Figs. 2D and 2G). However, in a subgroup of patients with loss of MMR protein expression (n=88), a high number of CDS+ T-lymphocytes was not associated with a better DSS or longer DFS (p=0.277 and p=0.208, respectively) (Figs. 2E and 2H).

- - - � f <:::r- ::i,_ � ;:i n, � � n, C/) � l l � cn "'S :=::- cn l:T' �::i ,Q., � � � g � � � ::i:: ;!l. -r;:: � � ti:.

- t"-< c:, -t - - .

§{ ::i,. � � � § ,.... n � :::· �

� [ � N g. n en

�

MLH1 I Negative Positive p-value

MSH2 Negative Positive p-value

MSH6 I Negative Positive p-value

MMR (l) I Negative Positive p-value

MMR (2) I Negative Positive p-value

MMR (3) I Negative Positive p-value

HLA class 1 1 HLA-A loss expressio11 loss

53 (49.1 %) 55 (50.9%) 33 (31 .4%) 133 (39.0%) 208 (61.0%) 100 (31.3%) 0.064 0.973

36 (52.9%) 32 (47.1 %) 24 (37.5%) 153 (39.5%) 234 (60.5%) 109 (29.9%) 0.039 0.223

22 (51.2%) 21 (48.8%) 16 (37.2%) 167 (40.3%) 247 (59.7%) 117 (30.4%) 0.170 0.359

73 (47.1 %) 82 (52.9%) 49 (32.9%) 116 (38.3%) 187 (61 .7%) 84 (30.0%) 0.070 0.538

30 (57.7%) 22 (42.3%) 19 (37.3%) 159 (39.3%) 246 (60.7%) 114 (30.2%) 0.011 0.304

8 (66.7%) 4 (33.3%) 5 (41 .7%) 181 (40.7%) 264 (59.3%) 128 (30.8%) 0.082 0.527

1 Downregulation of HLA class I heavy chain; either HLA-A or HLA-B/C MMR (1): concurrent negative expression of at least 1 MMR protein MMR (2): concurrent negative expression of at least 2 MMR proteins MMR (3): concurrent negative expression of 3 MMR proteins Bold values signifiJ p< 0.05

HLA-BIC �2m

expressio11 loss expression oss expressio11

72 (68.6%) 42 (39.3%) 65 (60 .7%) 1 (0.9%) 105 (99.1 %) 220 (68.7%) 89 (26.4%) 248(73.6%) 4 (1 .2%) 325 (98.8%)

0.011 0.999

40 (62.5%) 26 (39.4%) 40 (60.6%) 0 (0%) 67 (100%) 256 (70.1 %) 108 (28.1 %) 276 (71.9%) 5 (1 .3%) 368 (98.7%)

0.064 0.999

27 (62.8%) 16 (38.1 %) 26 (61 .9%) 0 (0%) 41 (100%) 268 (69.6%) 118 (28.8%) 292 (71.2%) 5 (1.3%) 395 (98.7%)

0.208 0.999

100 (67.1 %) 57 (37.7%) 94 (62.3%) 1 (0.7%) 149 (99.3%) 196 (70.0%) 77 (25.5%) 225 (74.5%) 4 (1 .4%) 288 (98.6%)

0.007 0.666

32 (62.6%) 23 (44.2%) 29 (55.8%) 0 (0%) 52 (100%) 264 (69.8%) 111 (27.8%) 289 (72.2%) 5 (1.3%) 385 (98.7%)

0.014 0.999

7 (58.3%) 4 (33.3%) 8 (66.7%) 0 (0%) 12 (100%) 288 (69.2%) 130 (29.5%) 310 (70.5%) 5 (1.2%) 424 (98.8%)

0.755 0.999

Chapter 3

50

Discussion

In this large and well-documented cohort of EC patients, we determined if MMR protein expression (MLHl, MSH2 and MSH6) was associated with HLA class I expression. We are the first to show that loss of HLA class I more frequently occurs in EC with concurrent loss of MMR protein expression and is mainly caused by selective loss of the HLA-B/C antigen. The number of CDS+ T-lymphocytes was not related to MMR expression. However, a high number of intratumoral CDS+ T-lymphocytes was associated with improved survival in the total group EC in contrast to the subgroup of patients with loss of MMR protein expression. Our results indicate that selective loss of HLA class I antigens contributes to immune escape mechanisms in EC with loss of MMR protein expression.

In our cohort, negative expression of MLHl, MSH2 or MSH6 was present among EC patients in 23.S¾, 15.2% and 9.3%, respectively. In agreement with previous studies in EC, negative MLHl expression occurred more frequently than MSH2 and MSH.6·22-26 Furthermore, we showed that 33.5% of EC patients have loss of MMR protein expression indicating an aberrant MMR system in these cases. This finding is in agreement with the general knowledge of MSI being a feature of type I EC with frequencies around 20-43%.4-5,27

MSI can be either detected by DNA extraction, amplification, gel electrophoresis analysis or immunohistochemistry. Detection by immunohistochemistry is based on complete loss of protein product of MSI related genes. Studies have previously shown that loss of immunohistochemical staining of MMR proteins is related to the corresponding MMR gene mutation and is useful in detecting the MSI phenotype in EC.25•27•28 In the current study, immunohistochemistry expression of the most important proteins involved (MLHl, MSH2 and MSH6) was determined in order to estimate the frequency of an abnormal MMR system. A possible disadvantage of immunohistochemistry is the possibility for false-positive staining; certain missence mutations in MLHl and MSH6 can lead to the translation of a non-functioning protein that can still be detected by immunohistochemistry. However, expression of these proteins is weak or heterogeneous in the cancer tissue.29 Where we only analyzed stainings when at least two TMA cores were present, each containing >20% tumor tissue, we believe that these stainings were representative for the entire tumor.3° Furthermore, only totally negative immunostaining for MMR proteins was used to exclude false positive cases caused by missence mutations and partial methylation as much as possible.

Tumors with abnormal MMR function are believed to be more competent at elicitating an immune response as compared to tumors with a normal MMR system31 as was demonstrated in colorectal carcinoma.7•9•10 Similar results were shown in a study among endometrioid EC where the number of TIL and peritumoral lymphocytes was significantly increased in tumors with MSJ+ status.32 Despite these promising results, the latter study did not provide information on the activation status of TIL and peritumoral lymphocytes which is of importance in colorectal carcinoma.7 In our current study, the number of intratumoral CDS+ T-lymphocytes was not significantly different between


tumors with negative or positive MMR protein expression. A possible explanation for the difference between our results and previous results in colorectal carcinoma might be caused by differences in methods for determining abnormal MMR function, quantification of the number of CDS+ T-lymphocytes or the pathogenesis of colorectal cancer and EC.

In tumors with an aberrant MMR system a strong and specific cellular response can also function as a selective pressure favoring the outgrowth of tumor cell clones that have lost HLAclass I expression.8•14 By downregulation of HLAclass I, these tumor cells are therefore able to escape from recognition and destruction by the immune system12•13,

as was shown in colorectal carcinoma.14 We observed that loss of HLA class I expression was present in 46.4% of the tumors with loss of MMR protein expression compared to 3S.6% in tumors with expression of MLHl, MSH2 and MSH6, without reaching statistical significance. However, downregulation of HLA class I heavy chain was associated with concurrent negative expression of 2 MMR proteins. Furthermore, selective downregulation of HLA-B/C was associated with negative MLHl expression and tumors with negative expression of �1 MMR protein. These results imply an association between loss of MMR protein expression and HLA class I downregulation in EC.

A previous study showed that in HNPCC-related colorectal carcinoma, loss of HLA class I expression was predominately caused by loss of �2m expression (46.9%) in contrast to sporadic tumors where only one case (5.0%) showed loss of �2m expression. 14 In our cohort of sporadic EC, we showed that loss of �2m expression was a rare phenomenon (1.1 %) which is in agreement with numbers in sporadic colorectal carcinomas.7•14

It must be noticed that methods for determining HLA class I loss differ significantly between studies. In our study for example, absence of HLA class I was defined as complete loss of expression where other studies compare normal expression to downregulation (i.e. partial loss and total loss).

In contrast to colorectal cancer, results are conflicting on the association between an abnormal MMR system and survival in EC (reviewed in 5•11) . We observed that patients with expression of MMR proteins have a better DSS compared to patients with loss of MMR protein expression. As was previously shown, high numbers of intra-tumoral CDS+ T-lymphocytes are associated with a better DSS and a longer DFS.16 However, this was not observed in a subgroup of patients with loss of MMR protein expression. This can be explained by the fact that a fully assembled HLA class I molecule is essential for CDS+ T-lymphocytes to recognize and kill tumor cells and therefore prevent further tumor growth.12 In agreement, we observed that the number of CDS+ T-lymphocytes was significantly lower in tumors with loss of HLA class I expression compared to tumors with expression. Where we also showed that patients with loss of MMR protein expression more often have downregulation of HLA class I, CDS+ T-lymphocytes cannot exert its normal function in destroying tumor cells in these patients. In contrast, patients with MMR protein expression more frequently maintain a functional HLA class I molecule for recognition by CDS+ T-lymphocytes and high number of CDS+ T-lymphocytes therefore maintained its prognostic favorable role on DSS and DFS (both: p<0.001). These results imply that as in colorectal carcinoma, EC patients with an abnormal MMR

51

Chapter 3

52

system more frequently have downregulation of HLA class I which makes these tumors able to escape specific immune responses by CDB+ T-lymphocytes.

In summary, the current study aimed to gain more insight in escape mechanisms of endometrial carcinoma and investigated the frequencies of loss of MMR protein expression and related this to HLA class I expression. We showed that EC's with loss of MMR protein expression more frequently show loss of HLA class I (mainly HLA-B/C) compared to patients with expression of MMR proteins. Patients with loss of MMR protein expression have a worse DSS compared to patients with expression. Furthermore, in patients with loss of MMR protein expression, CDB+ T-lymphocytes do not have a positive influence on survival which might be caused by the lack of normal HLA class I molecules in these patients. Above-mentioned results imply that EC patients with abnormal MMR expression are susceptible for selective pressure by the immune system which causes downregulation of HLA class I and escape from normal tumor-specific immune responses.


References

1 . Jemal A, Siegel R, Xu J , Ward E . Cancer statistics, 2010. CA Cancer J Clin 2010;60:277-300. 2. Bokhman JV. Two pathogenetic types of endometrial carcinoma. Gynecol Oncol 1983;15:10-7. 3. McCluggage WG, Robboy SJ. Mesenchymal uterine tumors and adenomyosis. In: Robboy SJ,

Mutter GL, Prat J, Bentley RC, Russell P, Anderson MC, editors. Pathology of the female reproductive tract. Second edition. Churchill Livingstone Elsevier; 2009. p. 427-56.

4. Catasus L, Machin P, Matias-Guiu X, Prat J. Microsatellite instability in endometrial carcinomas: clinicopathologic correlations in a series of 42 cases. Hum Pathol 1998;29:1160-4.

5. Karamurzin Y, Rutgers JK. DNA mismatch repair deficiency in endometrial carcinoma. Int J Gynecol Pathol 2009;28:239-55.

6. Guastadisegni C, Colafranceschi M, Ottini L, Dogliotti E. Microsatellite instability as a marker of prognosis and response to therapy: a meta-analysis of colorectal cancer survival data. Eur J Cancer 2010;46:2788-98.

7. Dolcetti R, Viel A, Doglioni C, Russo A, Guidaboni M, Capozzi E, Vecchia to N, Macri E, Fornasarig M, Boiocchi M. High prevalence of activated intraepithelial cytotoxic T lymphocytes and increased neoplastic cell apoptosis in colorectal carcinomas with microsatellite instability. Am J Pathol 1999;154:1805-13.

8. Kloor M, Becker C, Benner A, Woerner SM, Gebert J, Ferrone S, Knebel Doeberitz M. Immunoselective pressure and human leukocyte antigen class I antigen machinery defects in microsatellite unstable colorectal cancers. Cancer Res 2005;65:6418-24.

9. Greenson JK, Huang SC, Herron C, Moreno V, Bonner JD, Tomsho LP, Ben Izhak 0, Cohen HI, Trougouboff P, Bejhar J, Sova Y, Pinchev M, et al. Pathologic predictors of microsatellite instability in colorectal cancer. Am J Surg Pathol 2009;33:126-33.

10. Smyrk TC, Watson P, Kaul K, Lynch HT. Tumor-infiltrating lymphocytes are a marker for microsatellite instability in colorectal carcinoma. Cancer 2001;91:2417-22.

11. Black D, Soslow RA, Levine DA, Tornos C, Chen SC, Hummer AJ, Bogomolniy F, Olvera N, Barakat RR, Boyd J. Clinicopathologic significance of defective DNA mismatch repair in endometrial carcinoma. J Clin Oncol 2006;24:1745-53.

12. Algarra I, Garcia-Lora A, Cabrera T, Ruiz-Cabello F, Garrido F. The selection of tumor variants with altered expression of classical and nonclassical MHC class I molecules: implications for tumor immune escape. Cancer Immunol lmmunother 2004;53:904-10.

13. Seliger B. Strategies of tumor immune evasion. BioDrugs 2005;19:347-54. 14. Dierssen JW, de Miranda NF, Ferrone S, van Puijenbroek M, Cornelisse CJ, Fleuren GJ, van

Wezel T, Morreau H. HNPCC versus sporadic microsatellite-unstable colon cancers follow different routes toward loss of HLA class I expression. BMC Cancer 2007;7:33.

15. Bijen CB, Bantema-Joppe EJ, de Jong RA, Leffers N, Mourits MJE, Eggink HF, van der Zee AGJ, Hollema H, de Bock GH, Nijman HW. The prognostic role of classical and nonclassical MHC class I expression in endometrial cancer. Int J Cancer 2010;126:1417-27.

16. de Jong RA, Leffers N, Boezen HM, ten Hoor KA, van der Zee AGJ, Hollema H, Nijman HW. Presence of tumor-infiltrating lymphocytes is an independent prognostic factor in type I and II endometrial cancer. Gynecol Oncol 2009;114:105-10.

17. Creutzberg CL, van Putten WL, Koper PC, Lybeert ML, Jobsen JJ, Warlam-Rodenhuis CC, de Winter KA, Lutgens LC, van den Bergh AC, Steen-Banasik E, Beerman H, van Lent M. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355:1404-11.

53

Chapter 3

54

18. Seitz C, Uchanska-Ziegler B, Zank A, Ziegler A. The monoclonal antibody HCA2 recognises a broadly shared epitope on selected classical as well as several non-classical HLA class I molecules. Mol Immunol 1998;35:819-827.

19. Sernee MF, Ploegh HL, Schust DJ. Why certain antibodies cross-react with HLA-A and HLAG: epitope mapping of two common MHC class I reagents. Mol Immunol 1998;35:177-188.

20. Perosa F, Luccarelli G, Prete M, Favoino E, Ferrone S, Dammacco F. Beta 2-microglobulin-free HLA class I heavy chain epitope mimicry by monoclonal antibody HC-10-specific peptide. J Immunol 2003;171:1918-1926.

21. Hutter H, Hammer A, Blaschitz A, Hartmann M, Ebbesen P, Dohr G, Ziegler A, UchanskaZiegler B. Expression of HLA class I molecules in human first trimester and term placenta trophoblast. Cell Tissue Res 1996;286:439-447.

22. Cohn DE, Frankel WL, Resnick KE, Zanagnolo VL, Copeland LJ, Hampel H, Kelbick N, Morrison CD, Fowler JM. Improved survival with an intact DNA mismatch repair system in endometrial cancer. Obstet Gynecol 2006;108:1208-15.

23. Esteller M, Levine R, Baylin SB, Ellenson LH, Herman JG. MLHl promoter hypermethylation is associated with the microsatellite instability phenotype in sporadic endometrial carcinomas. Oncogene 1998;17:2413-7.

24. Goodfellow PJ, Buttin BM, Herzog TJ, Rader JS, Gibb RK, Swisher E, Look K, Walls KC, Fan MY, Mutch DG. Prevalence of defective DNA mismatch repair and MSH6 mutation in an unselected series of endometrial cancers. Proc Natl Acad Sci U S A 2003;100:5908-13.

25 . Modica I, Soslow RA, Black D, Tornos C, Kauff N, Shia J. Utility of immunohistochemistry in predicting microsa tellite instability in endometrial carcinoma. Am J Surg Pa thol 2007;31 (5) :7 44-51.

26. Stefansson I, Akslen LA, MacDonald N, Ryan A, Das S, Jacobs IJ, Salvesen HB. Loss of hMSH2 and hMSH6 expression is frequent in sporadic endometrial ·carcinomas with microsatellite instability: a population-based study. Clin Cancer Res 2002;8:138-43.

27. Arabi H, Guan H, Kumar S, Cote M, Bandyopadhyay S, Bryant C, Shah J, Abdul-Karim FW, Munkarah AR, Ali-Fehrni R. Impact of microsatellite instability (MSI) on survival in high grade endometrial carcinoma. Gynecol Oncol 2009;113:153-8.

28. Niessen RC, Berends MJ, Wu Y, Sijmons RH, Hollema H, Ligtenberg MJ, de Walle HE, de Vries EG, Karrenbeld A, Buys CH, van der Zee AG, Hofstra RM, et al. Identification of mismatch repair gene mutations in young patients with colorectal cancer and in patients with multiple tumours associated with hereditary non-polyposis colorectal cancer. Gut 2006;55:1781-8.

29. Boland CR, Koi M, Chang DK, Carethers JM. The biochemical basis of microsatellite instability and abnormal immunohistochemistry and clinical behavior in Lynch syndrome: from bench to bedside. Fam Cancer 2008;7:41-52.

30. Kononen J, Bubendorf L, Kallionierni A, Barlund M, Schraml P, Leighton S, Torhorst J, Mihatsch MJ, Sauter G, Kallioniemi OP. Tissue rnicroarrays for high-throughput molecular profiling of tumor specimens. Nat Med 1998;4:844-7.

31. Linnebacher M, Gebert J, Rudy W, Woerner S, Yuan YP, Bork P, Knebel Doeberitz M. Frameshift peptide-derived T-cell epitopes: a source of novel tumor-specific antigens. Int J Cancer 2001;93:6-ll.

32. Shia J, Black D, Hummer AJ, Boyd J, Soslow RA. Routinely assessed morphological features correlate with microsatellite instability status in endornetrial cancer. Hurn Pathol 2008;39:116-25.

Chapter 4

Serum tryptophan and kynurenine concentrations

as parameters for indoleamine 2,3-dioxygenase

activity in patients with endometrial,

ovarian and vulvar cancer

Renske A de Jong

Hans W. Nijman

H. Marike Boezen

Marcel Volmer

Klaske A ten Hoar

Jasper Krijnen

Ate G.J. van der Zee

Harry Hollema

Ida P. Kema

International Journal of Gynecological Cancer

2011;21:1320-1327.

Serum tn;ptophan and kynurenine concentrations as parameters for indoleamine 2,3-dioxygenase activity in patients with endometrial, ovarian and vulvar cancer

Abstract

Objective Indoleamine 2,3-dioxygenase (IDO) suppresses the function of T-lymphocytes and is involved in immune escape of cancers. IDO catalyses the initial rate-limiting step in the degradation of the essential amino-acid tryptophan. In this study, we investigated cancer-induced IDO activity in sera of endometrial, ovarian and vulvar cancer patients.

Methods Concentrations of tryptophan and kynurenine were determined in pretreatment serum samples of patients with endometrial (n=41), ovarian (n=28) and vulvar cancer (n=40) and compared to 19 healthy female controls. In serum of a subgroup of endometrial (n=22), ovarian (n=21) and vulvar (n=21) cancer patients, tryptophan, kynurenine and the kyn/ trp ratio were determined at different time points: preoperative, at clinical remission and at time of diagnosis of recurrent disease. Analyses were performed by an automated on-line solid-phase extraction-liquid chromatographic-tandem mass spectrometric (XLC-MS/MS) method. IDO activity was estimated by calculating the kynurenine-to-tryptophan ratio (kyn/ trp ).

Results Kynurenine concentrations (umol/L) and the kyn/trp ratio (umol/mmol) were higher in preoperative serum of endometrial, ovarian and vulvar cancer patients compared to controls (all: p<0.001). Preoperative serum of ovarian cancer patients contained higher kynurenine concentrations (median: 2.53; IQR: 1.72-4.29) and a higher kyn/trp ratio (median: 39.3; IQR: 26.5-61.7) compared to serum collected at clinical remission (median: 2.02; IQR: 1.68-2.72, p=0.035) and (median: 29.9; IQR: 23.4-38.9, p=0.005), respectively.

Conclusions Patients with endometrial, ovarian and vulvar cancer have increased tryptophan degradation compared to controls resulting in higher serum kynurenine concentrations and a higher kyn/ trp ratio. Our results suggest that !DO-induced immune escape may play an important role in these gynecological cancers.

57

Chapter 4

58

Introduction

Indoleamine 2,3-dioxygenase (IDO; EC 1.13.11.52) is currently considered one of the most important mechanisms for cancers to escape from host immune surveillance. 100 expression occurs in a wide range of malignancies.1 Furthermore, 100 overexpression is associated with poor prognosis in patients with endometrial and ovarian cancer.2,3

100 is an intracellular enzyme involved in tryptophan degradation. Tryptophan is an essential amino acid used for protein and indole synthesis and is metabolized along the kynurenine pathway where several downstream metabolites such as kynurenine, 3-hydroxykynurenine and ultimately nicotinamide adenine dinucleotide (NAO+) are formed.4 Next to IDO, tryptophan degradation along the kynurenine pathway is also catalyzed by tryptophan 2,3-dioxygenase (TOO; ECl.13.11.11). TOO is almost entirely located in the liver and is induced by tryptophan, tyrosine, histidine, glucocorticoids and kynurenine.5 In contrast, 100 is induced by pro-inflammatory mediators, such as endotoxin and interferon-y (IFN-y)6 after which IDO expression occurs in several cell types; macrophages, dendritic and cancer cells.7

The immunosuppressive potential of 100 was originally described in murine pregnancies where it prevents allogeneic fetal rejection by maternal T-cell immunity.8 Next, it was shown that 100 can induce peripheral tolerance in autoimmune disorders, chronic infections7 and cancer.1 Using immunohistochemistry, a study by Uyttenhove et al. showed that most human tumors constitutively express 100.1 Two mechanisms have been proposed how 100 might suppress the function of T-lymphocytes as well as natural killer (NK) cells. First, local tryptophan depletion leads to cell cycle arrest in the Gl phase and susceptibility for apoptosis.9 Alternatively, due to tryptophan degradation, metabolites are released which are toxic for T-lymphocytes and NK cells.5,10 Studies in several solid malignancies demonstrated an inverse correlation between 100 overexpression and the number of tumor-infiltrating lymphocytes. 1-3,11

In serum, IDO activity can be estimated by the kynurenine-to-tryptophan ratio (kyn/ trp ).12,13 In patients with cancer, tryptophan degradation is enhanced, resulting in lower serum concentrations of tryptophan, higher kynurenine concentrations and a higher kyn/trp ratio.12-17 A high kyn/trp ratio, which reflects !DO-induced immune escape of cancers, is associated with more advanced disease in lung cancer13 and ovarian cancer.18

Furthermore, a high kyn/ trp ratio predicts for poor prognosis in patients with acute myeloid leukemia14 and melanoma.17

In this study, concentrations of tryptophan, kynurenine and the kyn/trp ratio were determined in preoperative serum of endometrial, ovarian and vulvar cancer patients. Correlations between serum concentrations and clinicopathological factors were examined. In addition, serum concentrations were compared at different time points during patients' course of disease in order to determine whether IDO activity reflects tumor burden.

Serum tn;ptophan and kynurenine concentrations as parameters for indoleamine 2,3-dio:n;genase activity in patients with endometrial, ovarian and vulvar cancer

Material and methods

Patients After informed consent, pretreatment and follow-up serum samples for research purposes are routinely obtained from all patients with gynecologic malignancies treated at the Department of Gynecologic Oncology of the University Medical Center Groningen (Groningen, the Netherlands). After collection, serum is stored at -80°C until analysis. In light of this study especially patients with higher stage of endometrial, ovarian and vulvar cancer, who had undergone primary surgical treatment at our department, were selected. In order to assess variation in serum concentrations during the course of disease, an additional group of patients were selected who had been diagnosed with recurrent disease after a period of clinical complete remission. Of these patients, serum samples at three different time points were collected: preoperative, at clinical complete remission (at least 3 months after adjuvant radiotherapy or chemotherapy) and at time of diagnosis of recurrent disease. As a control group, preoperative sera of 19 females with benign gynecologic conditions for which they were operated were used. In total, 109 patients were included in this study. Preoperative serum samples could be analyzed of 41 endometrial, 28 ovarian and 40 vulvar cancer patients. A subgroup of these patients (22 endometrial, 21 ovarian and 21 vulvar cancer patients), was suitable for serum analysis at different time points during patients' course of disease. All patients were staged according to the FIGO guidelines.19 Tumors were classified and graded by pathologists according to the World Health Organization (WHO) criteria.20

Institutional review board approval For the present study, all relevant clinicopathological data were retrieved from our computerized database and transferred into a separate, anonymized, password protected database. Patient identity was protected by study-specific, unique patient codes, which were only known to two dedicated data managers, who also have daily responsibility for the larger database. In case of uncertainties with respect to clinicopathological and follow-up data, the larger databases could only be checked through the data managers, thereby ascertaining the protection of patients' identity. Due to these procedures according to Dutch law no further patient or institutional review board approval was needed (http://www.federa.org/).

Analytical procedures Tryptophan and kynurenine concentrations were determined by an automated on-line solid-phase extraction-liquid chromatographic-tandem mass spectrometric (XLC-MS/ MS) method with deuterated internal standards as described previously.4 Serum albumin concentrations were determined (as a measurement of protein status) using turbidimetry during routine blood examination at similar time points; reference range concentrations are between 38 and 49 g/L.

Statistics All continuous variables were checked for normality of the distribution using P-P plots. In case of skewed distributions, the median and interquartile ranges (IQR, 25th to 75th

59

Chapter 4

60

percentile) were presented. In order to test whether serum concentrations of tryptophan, kynurenine and the kyn/ trp ratio were significantly different between groups or time-points, Mann-Whitney U test and Wilcoxon signed-rank test were used, respectively. Spearman rank correlation analyses were used to determine correlations between continuous variables. All tests were performed two-sided and p-values <0.05 were considered to be statistically significant. Analyses were performed using the software package, version 16.0 for Windows (SPSS Inc. Chicago Illinois, USA).

Results

Comparison of tryptophan, kynurenine and the kynltrp ratio between patients and controls Patients Clinicopathological characteristics of endometrial, ovarian and vulvar cancer patients are summarized in Table 1. The majority of the selected patients were diagnosed with advanced stage of disease (FIGO stage III/IV) (endometrial cancer: 56.1 %; ovarian cancer: 89.3%; vulvar cancer: 77.5%). The median follow-up period was 3.3 years (IQR: 1.7-5.7). During this period, 50 patients (45.9%) died as a result of gynecological cancer; 12 endometrial cancer patients (29.3%), 19 ovarian cancer patients (67.9%) and 19 vulvar cancer patients (48.7%). Healthy controls were younger (median age: 60 years; IQR: 47-69) compared to vulvar cancer patients (median age: 70 years; IQR: 61-75) (p=0.004). No age differences were observed between healthy controls and endometrial cancer

Table 1. Clinicopathological characteristics of endometrial, ovarian and vulvar cancer patients.

Endometrial cancer Ovarian cancer Vulvar cancer (n=41) (N=28) (N=40)

Tumor type Endometrioid carcinoma 35 (85.4%) 1 (3.6%) 0 (0%) Clear cell carcinoma 3 (7.3%) 2 (7.1 %) 0 (0%) Serous papillary carcinoma 3 (7.3%) 19 (67.9%) 0 (0%) Mucinous carcinoma 0 (0%) 4 (14.3%) 0 (0%) Squamous cell carcinoma 0 (0%) 0 (0%) 41 (100%) Other 0 (0%) 2 (7.1 %) 0 (0%)

n,mor grade Grade 1 12 (29.3%) 1 (3.8%) 15 (37.5%) Grade 2 11 (26.8%) 7 (26.9%) 21 (52.5%) Grade 3 18 (43.9%) 17 (65 .5%) 4 (10.0%) Undifferentiated 0 (0%) 1 (3.8%) 0 (0%)

FIGO stage Stage I 7 (17.1 %) 1 (3 .6%) 3 (7.5%) Stage II 11 (26.8%) 2 (7.1 %) 6 (15.0%) Stage III 21 (51 . 2%) 19 (67.9%) 22 (55.0%) Stage N 2 (4.9%) 6 (21.4%) 9 (22.5%)

Serum tryptophan and ktJnurenine concentrations as parameters for indoleamine 2,3-dioxygenase activity in patients with endometrial, ovarian and vulvar cancer

patients (median: 65 years; IQR: 56-74) and ovarian cancer patients (median: 61 years; IQR: 44-72).

Serum concentrations of tryptophan, kynurenine and the kyn/trp ratio Median concentrations of tryptophan, kynurerune and the kyn/ trp ratio are presented in Table 2. Preoperative kynurerune concentrations were higher in endometrial (p<0.001), ovarian (p<0.001) and vulvar cancer patients (p<0.001) compared to controls (Figure 1). The preoperative kyn/trp ratio was higher in endometrial (p<0.001), ovarian (p<0.001) and vulvar cancer patients (p<0.001) compared to controls. Tryptophan concentrations did not significantly differ between patients and controls. A higher preoperative kyn/ trp ratio was found in ovarian cancer patients compared to endometrial cancer patients (p=0.029). Furthermore, preoperative serum of ovarian cancer patients contained higher kynurenine concentrations and a higher kyn/ trp ratio compared to vulvar cancer patients (p=0.029 and p=0.009, respectively). Serum tryptophan and kynurerune concentrations were not significantly different between endometrial and vulvar cancer patients.

Table 2. Preoperative serum concentrations of tn;ptophan, kt;nurenine and the kyn/trp ratio in endometrial, ovarian and vulvar cancer patients.

Endometrial cancer (n=41) Ovarian cancer (n=28) Vulvar cancer (n=40)

Controls (n=19)

n.s ,-----,

Tryptophan (umol/L) Median (IQR)

65.6 (58.3-73. 7) 67.3 (52.5-78.5) 67.7 (55.9-73.8)

64.0 (53.2-71.8)

Figure 1. Preoperative serum concentrations of tryptophan, ktprnrenine and the kt;n/trp ratio in endometrial, ovarian and vulvar cancer patients compared to healthy controls; significant differences between groups are indicated by p-value, non-significant differences are indicated by 'n.s. '

Kynurenine (umol/L) Kyn/Trp (umolhnmol) Median (IQR) Median (IQR)

2.00 (1. 26-2.54) 2.37 (1.47-4.18) 1 .78 (1 .46- 2.18)

1 .16 (0.91-1.33)

p<Q001

p<D001

p<0001 ,-----,

p<0.001

p<0.001

28.8 (22. 7-36.4) 37.2 (25.9-62.1) 27.9 (22.7-35. 2)

19.0 (15.5- 22.0)

� � pa0.029 p:0 009

61

Chapter 4

62

Association of tryptophan, kynurenine and the kynltrp ratio with clinical parameters Preoperative serum of ovarian cancer patients with advanced stage of disease (FIGO stage III/IV) contained significantly lower tryptophan concentrations (umol/L) (median: 66.6; IQR: 51.8-74.8) compared to patients with early stage of disease (FIGO stage 1/11) (median: 88.6; IQR: 68.2-116.3) (p=0.046). Furthermore, serum of advanced stage ovarian cancer contained a higher kyn/trp ratio (umol/mmol) (median: 43.5; IQR: 28.8-62.5) compared to early stage ovarian cancer (median: 19.7; IQR: 14.5-26.1) (p=0.021). Kynurenine concentrations were not significantly different between early and advanced stage ovarian cancer. In preoperative serum of vulvar cancer patients with advanced stage of disease (FIGO stage III/IV) lower concentrations of tryptophan (umol/L) (median: 60.3; IQR: 55.5-71.2) were detected compared to patients with early stage of disease (FIGO stage 1/11) (median: 76.2; IQR: 71.3-79.6) (p=0.006). Concentrations of kynurenine and the kyn/trp ratio were not significantly different between early and advanced stage vulvar cancer. In endometrial cancer patients, no significant differences were found between early and advanced stage of disease. No associations were found between tumor type, tumor grade and concentrations of tryptophan, kynurenine and the kyn/ trp ratio. A weak correlation was found between a person's age (cancer patients and controls) and the kyn/trp ratio (r5=0.243; p=0.006). No significant correlation was found between a person's age and tryptophan concentrations or kynurenine concentrations.

Albumin concentrations In preoperative serum of controls, albumin concentrations positively correlated with tryptophan concentrations (rs=0.634; p=0.027) but not with kynurenine concentrations or the kyn/ trp ratio. In preoperative serum of ovarian cancer patients, albumin concentrations positively correlated with tryptophan concentrations (rs=0.615; p=0.003) but not with kynurenine concentrations or the kyn/ trp ratio. In endometrial and vulvar cancer patients, no correlations were found between albumin concentrations and concentrations of tryptophan, kynurenine and the kyn/ trp ratio.

Variation of serum tryptoyhan, kynurenine and the kynltrp ratio during patients' course of disease Patients In serum of endometrial (n=22), ovarian (n=21) and vulvar (n=21) cancer patients, concentrations of tryptophan, kynurenine and the kyn/ trp ratio were compared between time points during patients' course of disease. Clinicopathological characteristics of this subgroup are summarized in Table 3.

Serum concentrations of tryptophan, kynurenine and the kyn/trp ratio Preoperative serum of ovarian cancer patients contained higher concentrations of kynurenine (umol/L) (median: 2.53; IQR: 1.72-4.29) and a higher kyn/trp ratio (umol/ mmol) (median: 39.3; IQR: 26.5-61.7) compared to serum collected at clinical remission (median: 2.02; IQR: 1.68-2.72, p=0.035) and (median: 29.9; IQR: 23.4-38.9, p=0.005),

Serum tryptophan and kynurenine concentrations as parameters for indoleamine 2,3-dioxygenase activity in patients with endometrial, ovarian and vulvar cancer

respectively (Figure 2). No significant differences between time points were found in serum of endometrial and vulvar cancer patients.

Table 3. Clinicopathological characteristics of a subgroup of endometrial, ovarian and vulvar cancer patients in which serum was analyzed at three different time points during patients' course of disease.

Endometrial cancer (N=22)

nunor type Endometrioid carcinoma 19 (86.4%) Clear cell carcinoma 2 (9.1 %) Serous papillary carcinoma 1 (4.5%) Mucinous carcinoma 0 (0%) Squamous cell carcinoma 0 (0%) Other 0 (0%)

Tumor grade Grade 1 10 (45.5%) Grade 2 7 (31 .8%) Grade 3 5 (22.7%) Undifferentiated 0 (0%)

FICO stage Stage I 8 (36.4%) Stage II 4 (18.2%) Stage III 9 (40.9%) Stage IV 1 (4.5%)

............

Figure 2. Serum concentrations of J tryptophan, kynurenine and kyn/trp at ,. .. different time points in ovarian cancer I patients; significant differences between time points are indicated by p-value, non-significant differences are indicated by 'n.s. '

Ovarian cancer Vulvar cancer (N=21) (N=21)

1 (4.8%) 0(0%) 2 (9.5%) 0 (0%) 15 (71 .4%) 0 (0%) 2 (9.5%) 0 (0%) 0 (0%) 21 (100%) 1 (4.8%) 0 (0%)

1 (4.8%) 12 (57.1 %) 7 (33.3%) 9 (42.9%) 13 (61 .9%) 0 (0%) 0 (0%) 0 (0%)

1 (4.8%) 4 (19.0%) 2 (9.5%) 6 (28.6%) 14 (66.7%) 9 (42.9%) 4 (19.0%) 2 (9.5%)

P"'0,035

J)"0.005

-

63

Chapter 4

64

Discussion

In this study, we show that preoperative serum concentrations of kynurenine and the kyn/trp ratio are higher in patients with endometrial, ovarian and vulvar cancer compared to controls. Preoperative serum of ovarian cancer patients contained higher kynurenine concentrations and a higher kyn/trp ratio compared to serum collected at clinical remission. Our results support the hypothesis that patients with endometrial, ovarian and vulvar cancer have increased tryptophan degradation, caused by cancerinduced IDO activity.

Studies in hematological cancer, melanoma and lung cancer indicated that cancer-induced IDO activity is increased as shown by an increased serum kyn/trp ratio.13·15,17

Supporting this theory, our data demonstrated higher kynurenine concentrations and a higher kyn/ trp ratio in preoperative serum of endometrial, ovarian and vulvar cancer patients compared to controls. To date, two studies have been published analyzing IDO activity in serum of gynecological cancer patients. The first study consisted of 20 gynecological cancer patients of different origins; ovarian cancer (n=ll), corpus carcinoma (n=S), tuba carcinoma (n=l), vulva carcinoma (n=l) and sarcoma uteri (n=l). In this population, tryptophan concentrations were lower in patients compared to controls. K ynurenine and kyn/ trp levels were not different between groups. A possible explanation for these contrasting results is size and composition of study populations; where our study analyzed preoperative serum of 41 endometrial, 28 ovarian and 40 vulvar cancer patients they analyzed serum of 20 patients (13 preoperative and 7 at recurrent disease).12 The second study was conducted in 20 ovarian cancer patients demonstrating lower tryptophan and a higher kyn/ trp ratio in patients compared to controls but kynurenine concentrations did not differ between groups.18 In contrast to our study, serum was analyzed after surgical treatment and before chemotherapy where we compared preoperative serum of ovarian cancer patients to controls which is probably a better reflection of active cancer.

We demonstrate that ovarian cancer patients with advanced disease had lower tryptophan concentrations and a higher kyn/ trp ratio compared to patients with early stage of disease. In addition, tryptophan concentrations were lower in advanced stage vulvar cancer compared to early stage vulvar cancer. A similar result was seen in patients with lung cancer and ovarian cancer where lower tryptophan concentrations and a higher kyn/ trp ratio were present in patients with advanced stage of disease. 13,18 These results indicate that IDO activity is enhanced in patients with a larger tumor burden.

To our knowledge, it has not yet been clarified whether tryptophan, kynurenine and the kyn/trp ratio change in one individual during the course of malignant disease. To this end, serum samples of individual patients at different time points were collected; "preoperative" for major tumorload, "remission" for clinical remission after primary treatment and "recurrent disease" for minor or major tumorload after a period of clinical remission. In ovarian cancer patients, kynurenine concentrations and the kyn/ trp ratio were higher in "preoperative" compared to "remission" serum. This result was not observed in patients with endometrial and vulvar cancer. A possible explanation

Serum tryptophan and kynurenine concentrations as parameters for indoleamine 2,3-dioxygenase activity in patients with endometrial, ovarian and vulvar cancer

for this different finding could be that ovarian cancer patients generally have a larger tumor mass19 which could result in higher preoperative IDO activity and a greater decline towards remission. Supporting this theory, we observed that serum of ovarian cancer patients contained higher kynurenine concentrations and a higher kyn/ trp ratio compared to endometrial and vulvar cancer patients. A trend towards a similar observation was found in one other study in gynecological cancer patients.12

A weak correlation was demonstrated between a person's age and the kyn/trp ratio (rs=0.243; p=0.006). This finding is in agreement with a study in healthy individuals, which supported the hypothesis that aging is associated with a change in immune activation especially of the T-cell/macrophage system. Due to an IFN-y dependent increase of IDO activity amongst the elderly, infections, autoimmune diseases and malignancies occur more often.21 No significant difference was found between the age of controls and endometrial and ovarian cancer patients. In addition, correlation between age and kyn/ trp ratio was quite weak and it is therefore unlikely that a difference in kynurenine concentrations and kyn/trp ratio between groups is caused by an age-dependent increase of IDO activity.

Albumin concentrations were determined and used as a measurement of protein status, which correlated with tryptophan concentrations in serum of controls (rs = 0.634; p=0.027) and ovarian cancer patients (rs=0.615; p=0.003). Similar results were found in patients with lung cancer13 and hematological malignancies.22 As a result of diminished nourishment in cancer patients, decreased albumin concentrations might cause a decline in tryptophan concentration. However, where changes in kynurenine are independent of this process and no significant correlations were found between albumin and the kyn/trp ratio, we believe that albumin changes have minor influence on the change of kyn/trp ratio (IDO activity) in our population.

Next to IDO, tryptophan degradation is catalyzed by TDO. However, there are essential differences; IDO is ubiquitously expressed throughout the body after induction by IFN-y, where TDO is almost exclusively located in the liver and is induced by glucocorticoids.5 Furthermore, TDO does not seem to play a role in immunological processes in contrast to IDO, which is known for its role in tumor-induced immune escape.23 The role of TDO in cancer patients is currently not clear and should be further investigated. However, we do not believe that TDO significantly contributes to an increased kyn/ trp ratio in gynaecological cancer patients based on the above-mentioned facts. Therefore, a cancer-induced increase of IDO activity is responsible for increased tryptophan degradation as observed in patients with gynaecological cancer. However, the question remains whether increased IDO activity in cancer patients is caused by spontaneous activity of tumor cells as part of their tumor-escape process as was suggested in a study by Uyttenhove et al. This study showed that tumor cell lines constitutively express IDO in the absence of IFN-y exposure.1 In contrast, other studies have demonstrated a correlation between IDO activity and neopterin concentrations in serum of cancer patients.12•16•18 Neopterin is produced by human macrophages and dendritic cell after activation by IFN-y but is generally absent in tumor cells and cell lines.12•24 Therefore, a raise in the kyn/ trp ratio accompanied by an increase in serum neopterin concen-

65

Chapter 4

66

tration suggests that IDO is primarily induced by immune cells. In addition, a recent study in ovarian cancer patients showed that higher kynurenine and kyn/trp ratio and lower tryptophan concentrations were associated with higher concentrations of several immune activation markers.18 These results suggest that IDO is induced by proinflammatory stimuli as part of an antitumor response instead of spontaneous activity of the tumor, although we can not exclude that to some extend the tumor cells itself can induce IDO activity.

In conclusion, our current data indicate that patients with endometrial, ovarian and vulvar cancer have a higher IDO activity compared to controls. In order to quantify an IDO-induced immune escape of cancers, one can determine tryptophan and kynurenine in serum of cancer patients. Its significance for predicting the course of malignant disease remains unclear. Currently, we are performing a study on tumor tissue in a large cohort of gynecological cancer patients in order to unravel the influence of local IDO activity on peripheral tryptophan depletion and prognosis.

Serum tryptophan and kynurenine concentrations as parameters for indoleamine 2,3-dioxi;genase activity in patients with endometrial, ovarian and vulvar cancer

References

1 . Uyttenhove C, Pilotte L, Theate I , et al. Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat Med 2003;9:1269-74.

2. Inaba T, Ino K, Kajiyama H, et al. Role of the immunosuppressive enzyme indoleamine 2,3-dioxygenase in the progression of ovarian carcinoma. Gynecol Oncol 2009;115:185-92.

3. Ino K, Yamamoto E, Shibata K, et al. Inverse correlation between tumoral indoleamine 2,3-dioxygenase expression and tumor-infiltrating lymphocytes in endometrial cancer: its association with disease progression and survival. Clin Cancer Res 2008;14:2310-7.

4. de Jong WH, Smit R, Bakker SJ, et al. Plasma tryptophan, kynurenine and 3-hydroxykynurenine measurement using automated on-line solid-phase extraction HPLC-tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci 2009;877:603-9.

5. Lob S, Konigsrainer A, Rammensee HG, et al. Inhibitors of indoleamine-2,3-dioxygenase for cancer therapy: can we see the wood for the trees? Nat Rev Cancer 2009;9:445-52.

6. Taylor MW, Feng GS. Relationship between interferon-gamma, indoleamine 2,3-dioxygenase, and tryptophan catabolism. FASEB J 1991;5:2516-22.

7. Mellor AL, Munn DH. IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nat Rev Immunol 2004;4:762-74.

8. Munn DH, Zhou M, Attwood JT, et al. Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 1998;281:1191-3.

9. Munn DH, Shafizadeh E, Attwood JT, et al. Inhibition of T cell proliferation by macrophage tryptophan catabolism. J Exp Med 1999;189:1363-72.

10. Della Chiesa M, Carlomagno S, Frumento G, et al. The tryptophan catabolite L-kynurenine inhibits the surface expression of NKp46- and NKG2D-activating receptors and regulates NKcell function. Blood 2006;108:4118-25.

11 . Brandacher G, Perathoner A, Ladumer R, et al. Prognostic value of indoleamine 2,3-dioxygenase expression in colorectal cancer: effect on tumor-infiltrating T cells. Clin Cancer Res 2006;12:1144-51.

12. Schroecksnadel K, Winkler C, Fuith LC, et al. Tryptophan degradation in patients with gynecological cancer correlates with immune activation. Cancer Lett 2005;223:323-9.

13. Suzuki Y, Suda T, Furuhashi K, et al. Increased serum kynurenine/tryptophan ratio correlates with disease progression in lung cancer. Lung Cancer 2010;67:361-5.

14. Corm S, Berthon C, Imbenotte M, et al. Indoleamine 2,3-dioxygenase activity of acute myeloid leukemia cells can be measured from patients' sera by HPLC and is inducible by IFN-gamma. Leuk Res 2009;33:490-4.

15. Hoshi M, Ito H, Fujigaki H, et al. Indoleamine 2,3-dioxygenase is highly expressed in human adult T-cell leukemia/lymphoma and chemotherapy changes tryptophan catabolism in serum and reduced activity. Leuk Res 2009;33:39-45.

16. Huang A, Fuchs D, Widner B, et al. Serum tryptophan decrease correlates with immune activation and impaired quality of life in colorectal cancer. Br J Cancer 2002;86:1691-6.

17. Weinlich G, Murr C, Richardsen L, et al. Decreased serum tryptophan concentration predicts poor prognosis in malignant melanoma patients. Dermatology 2007;214:8-14.

18. Sperner-Unterweger B, Neurauter G, Klieber M, et al. Enhanced tryptophan degradation in patients with ovarian carcinoma correlates with several serum soluble immune activation markers. Immunobiology 2010;216:296-301.

67

Chapter 4

68

19. Cancer Comittee of the I nternational Federation of Gynaecology and Obstetrics (1986). Staging announcement: FIGO Cancer Committee. Gynecol Oncol 2008;25:383-385.

20. Tavassoli FA, Devilee P. World Health Organization: Tumours of the Breast and Female Genital Organs (WHO Classification of Tumours). 1 ed. Lyon, IARCPress-WHO; 2003.

21. Frick B, Schroecksnadel K, Neurauter G, et al. Increasing production of homocysteine and neopterin and degradation of tryptophan with older age. Clin Biochem 2004;37:684-7.

22. Denz H, Orth B, Weiss G, et al. Weight loss in patients with hematological neoplasias is associated with immune system stimulation. Clin I nvestig 1993;71:37-41 .

23. Thackray SJ, Mowat CG, Chapman SK. Exploring the mechanism of tryptophan 2,3-dioxygenase. Biochem Soc Trans 2008;36:1120-3.

24. Huber C, Batchelor JR, Fuchs D, et al. Immune response-associated production of neopterin. Release from macrophages primarily under control of interferon-gamma. J Exp Med 1984;160:310-6.

Chapter 5

Immunosuppressive mechanism of indoleamine

2,3-dioxygenase and its prognostic role in

endometrial carcinoma

Renske A. de Jong

Ido P. Kema

Annemarie Boerma

H. Marike Boezen

Johannes J.L. van der Want

Marloes J.M. Gooden

Harry Hollema

Hans W. Nijman

This chapter is partly based on:

Prognostic role of indoleamine

2,3-dioxi;genase in endometrial carcinoma.

De Jong RA, Kema IP, Boenna A, Boezen HM,

van der Want ]JL, Gooden MJM, Hollema H, Nijman HW.

Gynecol Oncol 2012;126:474-80.

Immunosuppressive mechanism of indoleamine 2,3-dioxygenase and its prognostic role in endometrial carcinoma

Abstract

Objective Indoleamine-2,3-dioxygenase (IDO) catalyzes the degradation of the essential aminoacid tryptophan. In this way, IDO suppresses the function of T-lymphocytes and is an important immune escape mechanism for cancer. It is to be expected that IDO influences prognosis of cancer patients. This study investigated the suppressive mechanism of IDO and the prognostic influence of IDO expression in EC patients.

Methods The structural localization of IDO was analyzed in a human cancer cell line treated with different amounts of IFN-y (0, 250 or 750 IU /ml) by immunohistochemistry and electron microscopy. Extra- and intracellular concentrations of tryptophan and kynurenine were measured by XLC-MS/MS. A tissue microarray containing primary EC tissue of 355 patients was used to evaluate IDO expression. IDO expression was associated with clinicopathological characteristics, survival and previously determined numbers of intratumoral CD8+ and Foxp3+ T-lymphocytes.

Results An IFN-y dependent increase of IDO expression was observed in cancer cells. Electron microscopy showed that IDO was exclusively localized below the cellular membrane. Extracellular concentrations of tryptophan decreased dramatically (factor 339) after treatment with IFN-y while kynurenine concentrations increased (factor 25). Intracellular changes were less significant (factor 2). IDOhigh expression was associated with lower numbers of intratumoral CD8+ T-lymphocytes (p=0.031). Next to well-known prognostic parameters, IDOhigh expression was independently associated with poor disease specific survival in the general cohort of EC patients and among patients with early stage EC.

Conclusion Our results might help to understand why T-lymphocytes are more prone for apoptosis compared to cancer cells. Furthermore, IDOhigh expression was associated with poor survival of EC patients. This provides evidence that further research into the use of !DO-blocking agents in cancer treatment is valid where it might be a promising new therapeutic strategy.

71

Chapter 5

72

Introduction

Increased production of indoleamine-2,3-dioxygenase (IDO; EC 1.13.11.52) by tumor cells is an important cancer-induced immune escape mechanism.I IDO is an intracellular enzyme initiating the first and rate-limiting steps in the breakdown of the essential amino acid tryptophan along the kynurenine pathway. Here, tryptophan is catabolized into kynurenine which is ultimately converted to nicotinamide adenine dinucleotide (NAD).2 IDO has an immunosuppressive potential which was originally described in murine models where IDO prevented allogeneic fetal rejection.3 Furthermore, IDO has a role in creating peripheral tolerance in autoimmune disorders and chronic infections.4

In response to endotoxin release and IFN-y secretion by activated T-cells, IDO is upregulated in several cell types; macrophages4, dendritic cells5 and various human cancers.I Cancer tissue comprises higher IDO levels compared to normal tissue.6 Studies on several solid tumors have shown that high IDO expression has a negative influence on survival.6•11

The immunosuppressive influence of IDO is exerted by suppression of effector T-lymphocytes and natural killer (NK) cells. Several mechanisms have been proposed. First, local tryptophan depletion leads to cell cycle arrest in the Gl phase and increased susceptibility for apoptosis of T-lymphocytes.12 Second, metabolites are released due to tryptophan breakdown which are directly toxic to cytotoxic T-lymphocytes and NK cells.13·I5 Third, IDO has the ability to convert naive T cells to differentiate into immunosuppressive regulatory T cells (Treg).I6·IB In several tumor types, an inverse correlation between IDO expression and the number of tumor-infiltrating T-lymphocytes was discovered. I,7,s,I9 It appears that especially T-lymphocytes are sensitive for tryptophan depletion. However, it is currently not clear why specifically T-lymphocytes are affected in contrast to tumor cells which do not undergo growth arrest or apoptosis.

!DO-induced suppression of T-lymphocytes can be reversed by 1-methyltryptophan (1-MT) which is a pharmacological inhibitor of IDO. When combining 1-MT with chemotherapy, a more potent anti-tumor effect has been observed compared to the use of both the single agents separately as tested in mouse models.20,2I These results suggest a potential role for anti-IDO agents in developing new therapeutic strategies for cancer treatment. This is an interesting new approach to optimize treatment and prognosis of, for example, endometrial carcinoma (EC) patients. EC is the most common gynecological cancer and the fourth most common cancer in women in the western world.22 Although overall survival is quite good among patients with early stage of disease; 87% for stage I and 76% for stage II, this declines with advanced disease (stage ill: 57%, stage N: 18%).23 Over the last few years, little progress has been made in improving treatment and prognosis of EC patients and therefore, new therapeutic strategies are needed.

In this study, the prognostic role of IDO was explored in a large and well-documented cohort of patients with EC. Furthermore, the study aimed to further unravel the mechanism that involves IDO expression and subsequently causes suppression of T-lymphocytes whereas cancer cells remain unaffected.


Methods

Cell lines and culture Human ovarian cancer cell line SKOV3 was maintained in DMEM high glucose medium supplemented with 10% heat inactivated fetal calf serum under standard incubator conditions (humidified atmosphere, 95% air, 5% CO2, 37 °C). Cells were seeded in cell culture flasks (Corning Incorporated) or poly-L-lysine coated 8-well LabTek chamber slides (Nunc, Inc., Naperville, USA), 15.000 cells/ chamber. After 24 h, different amounts of IFN-y (0, 250 or 750 IU/ml) were added to the medium and cells were grown for 48 h. Recombinant human IFN-y was purchased from Boehringer Ingelheim (Alkmaar, The Netherlands). As a positive control, 293 HEK (human embryonic kidney) cells transfected with adenovirus carrying IDO gene were used (kindly provided by Dr. L.E. Deelman, Department of Clinical Pharmacology, University Medical Center Groningen (UMCG), University of Groningen, The Netherlands). Medium was removed and the cell pellet was pre-fixed with Unifix (buffered formalin). Next, cell pellets were fixed and put in 20 mL PreservCyt Solution8 (methanol based) and put in The Cellient'" Automated Cell Block System (Hologic) which paraffin-embedded the cell material. For immunohistochemistry staining, 3 µm sections were cut.

Measurement of tryptophan and kynurenine concentrations in cell culture Tryptophan and kynurenine concentrations were determined in cell culture supernatants and cell lysates by an automated on-line solid-phase extraction-liquid chromatographic-tandem mass spectrometric (XLC-MS/MS) method with deuterated internal standards as described previously.2 In this way, tryptophan and kynurenine concentrations were calculated per mg protein. Analyses were performed six fold for cells treated under each condition; 0 IFN-y (IDO negative) and 750 IFN-y (IDO positive).

Western Blot Cells were lysed in lysis buffer (50 mM Tris/Cl,5 mm EDTA,150 mM NaCl,0.5% Triton X-100,pH 7.5). Protein concentration was determined using the BioRad DC assay. Sample buffer was added to the cell lysates at the ratio 1:4 (v /v) and boiled for 4 min. Lysates containing equal amounts of protein (25 µg) were used for all experiments. Protein fractions of whole cell lysates were separated by electrophoresis on 10% polyacrylamide gels in the presence of SDS and transferred onto activated PVDF membranes (Millipore BV, Etten-Leur, The Netherlands). Membranes were blocked with blocking buffer (5% dry milk in PBS containing 0.1 % Tween-20 (PBS-T)) for 1 h and incubated with primary antibody ( dilution: 1 :250), overnight at 4 °C. A mouse monoclonal antibody recognizing IDO (anti-indoleamine 2,3-dioxygenase, clone 10.1; Millipore (Chemicon)) was used as primary antibody for all described experiments. Membranes were washed 3 times for 10 min with PBS-T and incubated for 2 h at room temperature in goat-anti-mouse IgGalkaline phosphatase conjugated secondary antibody (Southern Biotec,dilution 1:5000 in TBS-T). After again three washing steps the membranes were developed by conventional method with nitroblue tetrazolium (NBT) / 5-bromo-4-chloro-3-inolyphosphate (BCIP) (Southern Biotech) in 100 mM Tris-HCI, 150 mM NaCl, and 5 mM MgC12 (pH 9.5)

73

Chapter 5

74

Patients and tissue samples From 1984 onwards clinicopathological parameters and follow-up data (Table 1) of all patients referred to the Department of Gynecological Oncology of the UMCG are prospectively collected during standard treatment and follow-up and stored in a computerized registration database. Tissue samples of these patients are maintained in the tissue storage system of the Department of Pathology of the UMCG. For the present study, tissue material and data of EC patients were used as previously reported by our group.24 Patients had been consecutively selected when treated with primary surgery in the UMCG between 1984 and 2004. In total, 355 EC patients were included for this study. Patients were treated with total abdominal hysterectomy and a bilateral salpingo-oopherectomy (TAH-BSO). Pelvic and/ or para-aortic lymph node dissection was performed in case preoperative suspicion of cervical invasion (FIGO stage II)25 or high-grade histological features. Adjuvant radiotherapy was given if indicated.26 Staging was done according to the FIGO classifications and tumors had been graded and classified according to the WHO criteria27 by a pathologist specialized in gynecological oncology. Patients were followed up to 10 years with gradually increasing intervals. Follow-up data were completed until August 2010. All relevant data were retrieved from our computerized database and transferred into a separate, anonymized, password protected database. Patient identity was known only to two dedicated data managers. In case of uncertainties with respect to clinicopathological and follow-up data, the larger databases could be checked only through the data managers, thereby ascertaining the protection of patients' identity. Due to these procedures according to Dutch law no further patient or institutional review board approval was needed.

Immunohistochemistry Four micrometer sections were cut from previously constructed TMAs of formalin-fixed paraffin-embedded tumors24 and applied to 3-amino-propyl-triethoxy-silane coated glass slides (Sigma-Aldrich, Diesenhofen, Germany). Antibodies for CDS+ T-lymphocytes and Foxp3+ T-lymphocytes were applied and reported previously.24 Sections (cells and TMA) were dewaxed and rehydrated. For TMA, antigen retrieval was performed by microwave treatment; 15 min in EDTA buffer (pH 8). Endogenous peroxidase activity was blocked by incubating slides in a 0.3% Hp

2 solution for 30 min. Slides were

incubated with IDO antibody for 60 min at room temperature (dilution, 1:25 for cells and 1:100 for TMA)). RAMP0 (rabbit anti-mouse peroxidase-labeled, DAKO, Beverlee, Belgium, 1:100) and GARP0 (goat anti-rabbit, peroxidase-labeled, DAKO, 1:100) were used as secondary and tertiary antibodies respectively. Antigen-antibody reactions were visualized with 3,3' -diaminobenzidine (DAB) and sections were counterstained with hematoxylin. Two researchers independently scored immunohistochemistry stained slides at a double-headed microscope without prior knowledge of clinicopathological information. Staining was analyzed only when 2::2 cores were available, each containing 2::20% tumor tissue. In this way, the tissue samples resembled heterogeneity of whole tissue slides. IDO expression was assessed according to a semi-quantitative scale, based on extent and intensity of the cytoplasmic staining.10 Intensity was scored as O for absent, 1 for weak, 2 for positive or 3 for strong positive expression. The percentage of positive tumor cells was scored as O for 0-5%; 1 for 2::5-30%; 2 for 2::30-70% and 3 for 2::70-100%.


Table 1. Clinicopathological characteristics of 355 patients with EC

Characteristics Patients no (n = 355)

Age (years) Median (IQR) 64.0

FICO Stage Stage I 196 Stage II 58 Stage ill 78 Stage IV 24

Tumor grade Grade I 159 Grade II 92 Grade ill 98 Undifferentiated 6

Tumor type Endometrioid 306 Serous papillary 16 Clear cell 22 Mixed tumors 5 Undifferentiated 6

Myometrial invasion <50% 203 �50% 152

LVSI Negative 241 Positive 97 Missing 17

Treatment TAH+BSO 187 TAH+BSO+LND 155 Other 13

Lymph node status Negative 113 Positive 43 Missing 199

Radiotherapy No 148 Yes 207

Recurrence No recurrence 269 Recurrence: 86 Local 30 Pelvic region 10 Distant 46

LVSI = lymphvascular space involvement TAH+BSO = total abdominal hysterectomy with bilateral salpingo-oopherectomy LND = lymph node dissection 1 Percentages exclude missing cases

%1

56-73

55.2 16 .3 21 .7 6 .8

44.8 25 .9 27.8 1 . 7

86.2 4 .5 6 .2 6.2 1 .7

57 .2 42.8

71 .3 28 .7

52.7 43 .7 3 .6

72.4 27.6

41 . 7 58.3

75.8 24.2 34 .9 11 .6 53 .5

75

Chapter 5

76

The sum of scores was used to identify four categories of expression: IDO- (sum: 0-1), IDOl + (sum: 2-3), IDO2+ (sum: 4-5) and IDO3+ (sum: 6). Mean result of the cores was considered as definitive IDO expression in the tumor of this particular patient. Immunostaining for CDS+ T-lymphocytes and Foxp3+ T-lymphocytes was scored and reported previously.24

TissueFAXS Measurements of IDO intensity were obtained from 3 cell samples treated with 0, 250 and 750 IU /ml IFN-y on LabTek chamber slides. Individual cells were identified by hematoxylin nuclear staining at 200x magnification, and selected from 6 randomly chosen areas from 470 images. Automated image acquisition and multichannel immunohistochemistry measurements were obtained with a Zeiss Axiolmager Zl Microscope (Zeiss Oberkochen, Germany) equipped with a Pixlink PL-230 camera (PCO Ag Kelheim, Germany). For recording of the density HistoQuest 2.0.2 software (TissueGnostics, GmbH, Vienna) was used.

Electron Microscopy SKOV3 cells were passaged 24 h prior to treatment and plated onto Poly-L-lysine coated 8-well LabTek chamber slides. Cells were fixed in freshly prepared 4% paraformaldehyde in 0,1 M phosphate buffered saline (PBS), pH 7.6 for 30 min at 4°C and permeabilized with 0.05% Triton X-100 (Sigma) in PBS for 10 min at room temperature. Endogenous peroxidase activity was blocked by incubating the slides in a 0.3% H2O2

solution for 15 min. Slides were blocked with 5% bovine serum albumin (BSA) diluted in PBS for 60 min. The IDO antibody was used as primary antibody (dilution 1:250) in 1 % BSA in PBS. Slides were incubated for 1 h at room temperature in a humidified chamber. After washing, RAMP0 and GARP0 were used as above-mentioned. Antigenantibody reactions were visualized with DAB for 10 min at room temperature. The reaction was stopped by washing the cells in demi-Hp. Following labeling the DAB reaction precipitate was enhanced according the gold-substituted silver peroxidase method.28 The cells were kept on tissue slides, postfixed in 1 % osmium tetroxide in 0.lM cacodylate buffer supplemented with 1.5% potassium hexacyanoferrate for 15 min and subsequently dehydrated in graded series of ethanol and propylene oxide. Cells were embedded in epon according to routine procedures. After polymerization, epon sheets containing the cultured cells were mounted on prepolymerized epon blocks, trimmed to fit the ultramicrotome cutting. Ultrathin sections were contrasted with uranyl acetate and lead citrate and inspected in a CM 100, (FEI, Electron Optics, Eindhoven, the Netherlands). Images were collected using a Morada camera (Olympus, Sys, Muenster, Germany).

Statistical analysis All continuous variables were checked for normality of the distribution using P-P plots. In case of skewed distributions, the median and interquartile ranges (IQR, 25th to 75th percentile) were presented. Staining of IDO was dichotomized for further analysis; IDO- and IDOl + were considered IDQ1°w expression whereas IDO2+ and IDO3+ were considered IDQhigh expression. To test differences in tryptophan and kynurenine levels,


numbers of CDS+ T-lymphocytes and Foxp3+ T-lymphocytes between tissues samples with IDO'°'v and IDQhigh expression, Mann-Whitney U tests were used. Associations between IDO expression and clinicopathological characteristics (see Table 1) were tested using univariate logistic regression analyses. IDO expression was used as dependent and clinicopathological parameters as independent variables. Odds Ratios (ORs) and 95%-confidence intervals (95%-Cls) were calculated. Survival was expressed as disease free survival (DFS) and disease specific survival (DSS). DFS was defined as date of diagnosis until recurrent disease, metastasis, death due to EC or last date of follow-up. DSS was defined as date of diagnosis until death due to EC or the date of last follow-up. Univariate and multivariate Cox regression analyses were used to assess influence of clinicopathological characteristics, IDO expression and number of CDS+ T-lymphocytes on survival. These variables were used as independent and death of disease (DSS) or recurrent disease (DFS) were used as dependent variables. Hazard Ratios (HRs) and 95%-Cls were calculated. Variables with a p-value >0.05 in univariate analysis were not included in multivariate analysis which was performed using an "enter model". All tests were performed two-sided and p-values of <0.05 were considered statistically significant. Analyses were performed using the software package PASW, version 18.0 for Windows (SPSS Inc., Chicago Illinois, USA).

Results

Cell line experiments Western Blot analysis was used to investigate IDO expression in SKOV3 cell lines and to demonstrate selectivity of the antibody in the conditions applied. It could be demonstrated that IDO expression was IFN-y dependent. As shown in Figure 1, the monoclonal IDO antibody stained a clear and single band of approximately 42 kDa in the lysates of cells treated with 250 and 750 IFN-y IU /ml. A band of the same size was detected in 293 HEK cells. Staining was negative in lysates of untreated cells. To identify the fine structural localization of IDO in exposed cells, immunohistochemistry was performed. Cytoplasmic expression of IDO was observed in cancer cells after treatment with 250 and 750 IU/ml IFN-y whereas weak IDO expression was observed in cell lines without IFN-y treatment (Figure 3a-c). This observation was objectified by means of TissueFAXS analysis; cell samples treated with increasing concentrations of

Table 2. Intensity of !DO immunostaining measured by TissueFaxs analysis.

O IFN-y

250 IFN-y

750 IFN-y

control

DAB = 3,3'-diaminobenzidine BLUE = nuclear staining by haematoxylin

DAB

120.583 151.828 153.615

102.213

BLUE

194.267 166.851 166.624

193.151

multiplier

1.61 1 .10 1 .09

1 .89

77

Chapter 5

HlO

p=0.031 p-0.644

Figure 1 Figure 2

A

Figure 3

Figure 4

78


IFN-y showed an increased intensity of immunohistochemistry expression (DAB) while using hematoxylin (BLUE) as reference value (Table 2). In addition, electron microscopy images (Figure 3d-f) showed that the amount of IDO labelling was remarkably higher in cancer cells treated with IFN-y compared to the untreated cells. IDO labelling was exclusively localized at the level just below the cellular membrane, as was observed in randomly chosen samples. The number of labelled particles following O IFN-y exposure were around 5-8 particles per membrane surface outline, whereas after 250 and 750 IFN-y treatment the amount raised 5-7 fold to 25-35 fold immunoparticles, respectively.

Tryptophan and kynurenine concentrations in cell material To confirm that IDO expression in cell lines was associated with the presence of a functionally active enzyme, enzymatic assays by means of XLC-MS/MS were used. IDO activity was analyzed in terms of their ability to catabolise tryptophan to form kynurenine. Results are summarized in Table 3. Extracellular tryptophan concentrations were significantly lower in supernatant of cells treated with IFN-y compared to untreated cells (p=0.002). Kynurenine concentrations were significantly higher in supernatant of cells treated with IFN-y compared with untreated cell samples (p=0.002). After IFN-y treatment, extracellular concentrations of tryptophan and kynurenine changed with a factor 339 and 25, respectively. Intracellular tryptophan concentrations were significantly lower in lysates of cells treated with IFN-y compared to untreated cells (p=0.002). Kynurenine concentrations were significantly higher in lysates of cells treated with IFN-y compared with untreated cell samples. Intracellular concentrations of tryptophan and kynurenine both changed with a factor 2 after IFN-y treatment.

IDO expression in tumor tissue Patients Median age at time of diagnosis was 64 years (IQR: 56-73) (Table 1). The majority of the patients (71.5%) was diagnosed with early stage (I/II) of disease and had good or moderately differentiated tumors (70.7%). TAH-BSO was performed in 187 patients (52.7%) and 155 patients (43.7%) received more extended surgery with pelvic and/or para-aortic lymph node resection.

Table 3. Measurements of tryptophan and kynurenine concentration in supernatant (extracellular) and lysates (intracellular) of cell lines (umol/mg protein).

EXTRACELLULAR Tryptopltan (umol/mg protein)

O IFN-y 750 1FN-y multiplier p-value

Median (IQR) 132.37 (124.21-137.46) 0.39 (0.37-0.44) 339 0.002 Kynurenine (mnol/mg protein) Median (IQR) 3.40 (3.25-3.61) 84.97 (82.97-91.16) 25 0.002

INTRACELLULAR TnJptophan (umollmg protein) Median (IQR) 4.21 (3.74-5.37) Kynurenine (umollmg protein) Median (IQR) 4.47 (3.82-4.66)

1 .90 (1.60-2.14)

9.47 (9.11-11.50)

2

2

0.002

0.002

79

Chapter 5

80

IDO expression and tumor-infiltrating lymphocytes IDQhigh expression was observed in 18.1 % of primary EC tissue (Figure 4). Next, it was assessed whether the number of tumor-infiltrating CDS+ and Foxp3+ T-lyrnphocytes were associated with IDO expression. IDQhigh tumors had significantly lower number of CDS+ T-lymphocytes (median: 11; IQR 4-21) compared to IDO10w tumors (median: 16; IQR: 6-37, p=0.031) (Figure 2). The number of Foxp3+ T-lymphocytes did not differ between tumors with IDQhigh (median: 3; IQR: 1-8) and IDO10w expression (median: 4; IQR: 0-7, p=0.644).

IDO expression and clinicopathological parameters No association was observed between IDO expression and clinicopathological parameters (Table 4).

Table 4. Association between clinicopatlwlogical characteristics JOO expression.

IDO

Negative Positive

Age (continuous) Age >60 years 130/218 (59.6%) 26/50 (52.0%) Advanced disease (FIGO III/IV) 81/244 (33.2%) 14/54 (25.9%) Ttunor grade 3/ undifferentiated 78/244 (32.0%) 15/54 (27.8%) Non-endometrioid tumor type 35/244 (14.3%) 7/54 (13.0%) LVSI positive 70/234 (29.9%) 17/51 (33.3%) Myometrial invasion �50% 111/244 (45.5%) 21/54 (38.9%) Lymplz nodes positive 37/109 (33.9%) 5/23 (21.7%)

Survival analyses

ID01•igh expression

p-OR (95% CI) value

0.99 (0.97-1 .02) 0.414 0. 73 (0.40-1.36) 0.325 0.70 (0.36-1.37) 0.301 0.82 (0.43-1 .57) 0.548 0.89 (0.37-2.13) 0.792 1.17 (0.61-2.24) 0.631 0.76 (0.42-1 .39) 0.378 0.54 (0.19-1 .57) 0.259

Median time of follow-up was 5.3 years (IQR: 2.2-8.5). In 86 of 355 patients (24.2%) recurrent disease was diagnosed with median time to recurrence of 19.7 months (IQR: 11.0-36.3). During follow-up, 61 of 355 patients (17.2%) died as a result of EC. In univariate Cox regression analyses, well-known prognostic factors such as lymphvascular space involvement (LVSI), positive lymph nodes, advanced disease, �50% myometrial invasion, non-endometrioid tumor type, high tumor grade in addition to a low number of CDS+ T-lymphocytes were associated with a shorter DFS (Table 5). Covariates that independently predicted for shorter DFS on multivariate Cox regression analyses were: LVSI, advanced disease and low number of CDS+ T-lymphocytes. In a subgroup of early stage EC: positive LVSI, age >60 years and low number of CDS+

T-lymphocytes. In advanced stage EC: LVSI. In a subgroup of endometrioid EC, LVSI, advanced disease and a low number of CDS+ T-lymphocytes had a negative influence on DFS. IDO expression did not have an influence on DFS in the entire cohort of EC patients, nor in the subgroup analyses. Next, influence of clinicopathological characteristics and IDO expression on DSS was determined (Table 6). Overall, 5-year DSS rate was SO% in IDOlow patients compared to 72% in IDQhigh patients.

Table 5. COX regression analysis on Disease Free Survival (DFS)

Univariate analysis Multivariate analysis HR 95% CI p-value HR 95% CI p-value

All (n=355) Age 1.01 0.99-1.03 0.219 Age >60 years 1.40 0.89-2.22 0.146 Advanced disease (FIGO ill/IV) 2.93 1.92-4 .48 <0.001 1 .78 1 .08-2.94 0.024 Tumor grade 3/ undifferentiated 1 .85 1.20-2.86 0.006 1.00 0.48-2.06 0.990 Non-endometrioid tumor type 1 .98 1 .16-3 .37 0.012 1 .59 0 .68-3.70 0 .282 LVSI positive 4.12 2.69-6.40 <0.001 3.16 1 .91-5 .23 <0.001 Myometrial invasion �50% 2.27 1.48-3 .48 <0.001 1 .04 0 .51-2.13 0.916 Lymph nodes positive 3 .24 1 .86-5.67 <0.001 1.09 0.49-2.45 0 .828 IDQhigh expression 1 .30 0.77-2 .20 0.333 1 .18 0 .43-1.47 0 .468 CD8+ I-lymphocytes 0.96 0.94-0.98 <0.001 0.98 0.97-0.99 0.005

Early stage (n=254) Age 1 .03 1.00-1.05 0.067 Age >60 years 2 .36 1.15-4 .83 0.019 2 .82 1 .21-6.58 0.016 Tumor grade 3/ undifferentiated 1.85 0.97-3 .52 0.063 Non-endometrioid tumor type 1 .46 0.58-3.71 0.422 LVSI positive 3 .98 2 .17-7 .30 <0.001 4 .08 1 .95-8.53 <0.001 Myometrial invasion �50% 2.32 1 .29-4 .16 0.005 1.15 0.54-2.47 0 .719 IDQhigh expression 1 .71 0 .87-3.37 0 .123 1 .10 0 .27-1 .51 0 .301 CDS+ I-lymphocytes 0.97 0.95-0 .99 0.004 0.96 0.94-0 .99 0.003

Advanced stage (n=101) Age 1 .01 0.99-1.04 0.358 Age >60 years 1 .14 0.61-2.16 0.679 Tumor grade 3/ undifferentiated 1 .03 0.56-1.91 0.916 Non-endometrioid tumor type 1 .52 0.77-2.97 0 .227 LVSI positive 2.60 1.34-5.05 0.005 2.57 1.26-5 .22 0.009 Myometrial invasion �50% 1 .27 0.67-2 .43 0 .463 1 IDQhigh expression 1.18 0.49-2.81 0.716 0.65 0 .23-1 .85 0 .419 CDs+ T-lymphocytes 0.99 0.98-1.00 0 .106 0 .99 0 .98-1 .01 0 .227

Endometrioid (n=306) Age 1 .01 0.99-1 .03 0 .386 Age >60 years 1.33 0.80-2.22 0.267 Advanced disease (FIGO III/IV) 2.76 1.71-4 .46 <0.001 1 .84 1 .06-3.21 0.032 Tumor grade 3/ undifferentiated 1 . 74 1.00-3.00 0.049 0.95 0 .46-1 .99 0 .899 LVSI positive 3 .89 2 .41-6 .30 <0.001 2 .74 1 .58-4.76 <0.001 Myometrial invasion �50% 2.28 1 .42-3 .67 0.001 1 .26 0.58-2.72 0.558 Lymph nodes positive 3 .08 1 .62-5.86 0.001 1 .14 0 .44-2 .92 0.792 IDOh,gh expression 1.18 0.65-2 .13 0.594 0.82 0 .42-1 .60 0.559 CDS+ I-lymphocytes 0.98 0.97-0 .99 0.006 0.98 0.97-0 .99 0.014

Non-endometrioid (n=49) Age 1 .02 0 .98-1 .07 0 .286 Age >60 years 1 .60 0.56-4 .63 0.383 Advanced disease (FIGO ill/IV) 2.57 0.90-7.34 0 .077 LVSI positive 4.33 1 .39-13.48 0.012 1 .59 0.31-8.15 0.582 Myometrial invasion �50% 1 .90 0.70-5 .16 0 .205 Lymph nodes positive 3.40 1 .07-10 .80 0.038 1.51 0.40-5 . 72 0.544 IDQhigh expression 2 .34 0 .75-7.31 0 .143 0 .78 0 .16-3.87 0 .756 CDS+ I-lymphocytes 0.98 0 .96-1 .00 0.092 0 .99 0 .97-1 .01 0.257

Bold values signifiJ p < 0.05 LVSI = lymphvascular space involvement 1 Not included in multivariate analysis due to non significant value in univariate analysis. IDO expression and CDS+ T-lymphocytes were included in multivariate analysis irrespective of the p-value.

81

82

Table 6. COX regression analysis on disease specific survival (055)

All (11=355) Age (continuous) Age >60 years Advanced disease (FI CO III /IV) Tumor grade 3/ undifferentiated Non-endometrioid tumor type LV SI positive Myometrial invasion <!:50% Lymph nodes positive IDQhigh expression CDS+ I-lymphocytes

Early stage (n=254) Age (continuous) Age >60 years Tumor grade 3/ undifferentiated Non-endometrioid tumor type LV SI positive Myometrial invasion <!:50% IDOhigh expression CDS+ I-lymphocytes

Advanced stage (n=101) Age (continuous) Age >60 years Tumor grade 3/ undifferentiated Non-endometrioid tumor type LV SI positive Myometrial invasion <!:50% IDOhigh expression CDS+ I-lymphocytes

Endometrioid (n=306) Age (continuous) Age >60 years Advanced disease (FICO III /IV) Tumor grade 3/ undifferentiated LV SI positive Myometrial invasion <!:50% Lymph nodes positive IDOhigh expression CDS+ I-lymphocytes

Non-endometrioid (n=49) Age (continuous) Age >60 years Advanced disease (FI CO III/IV) LV SI positive Myometrial invasion <!:50% Lymph nodes positive IDOhigh expression CDS+ I-lymphocytes

Bold values signifiJ p < 0.05

Univariate analysis HR 95% CI

1.02 1.00-1.05 1 .00 0.57-1 .75 7.24 4.18-12.57 2.81 1 .70-4.64 3.74 2.19-6.39 4.93 2.91-8.37 3.77 2.18-6.55 3.36 1 .78-6.35 1.88 1.06-3.32 0.96 0.94-0.98

1.07 1.02-1 .13 2.23 0.60-8.23 2.21 0.83-5.90 2.19 0.63-7.60 4.87 1 .86-12.75 3.43 1 .33-8.85 3.65 1.37-9.72 0.94 0.89-0.99

1.02 1.00-1.05 1 .13 0.60-2.15 1.40 0.77-2.56 2.36 1 .28-4.35 2.21 1 .16-4.22 1.89 0.95-3 .75 2.02 0.96-4.24 0.97 0.95-0.99

1.01 0.99-1 .04 0.60 0.30-1.22 6.24 3.30-11.80 0.97 0.98-3.94 4.91 2.60-9.28 4.15 2.12-8.14 2.79 1.29-6.05 1 .68 0.83-3.41 0.96 0.94-0.99

1.05 1.00-1.10 2.51 0.82-7.66 5.95 1 .73-20.42 2.91 1 .10-7.72 2.28 0.88-5.95 4.16 1 .24-13.93 3.72 1 .39-9.91 0.97 0.94-1.00

LV5I = lymphvascular space involvement

p-value

0.056 0.996 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.030 <0.001

0.004 0.231 0.112 0.216 0.001 0.011 0.010 0.022

0.090 0.698 0.275 0.006 0.017 0.070 0.064 0.007

0.313 0.157 <0.001 0.055 <0.001 <0.001 0.009 0.147 0.005

0.040 0.106 0.005 0.032 0.091 0.021 0.009 0.044

Multivariate analysis HR 95% CI p-value

1 1 4.39 2.29-8.43 <0.001 0.84 0.33-2.14 0.715 2.45 1.34-4.47 0.004 1 .36 0.52-3.57 0.529 2.04 1.11-3.78 0.023 0.74 0.31-1 .76 0.491 1 .86 1 .01-3.42 0.047 0.97 0.95-0.99 0.004

1 .07 1 .02-1 .13 0.008 1 1 1 4.70 1 .64-13.50 0.004 1 .98 0.57-6.90 0.286 3.05 1 .08-8.44 0.033 0.94 0.89-1 .00 0.059

1 1 1 2.24 1 .12-4.51 0.023 1 .87 0.90-3.88 0.094 1 1 .29 0.56-2.99 0.548 0.98 0.96-1 .00 0.068

1 1 4.29 1 .96-9.38 <0.001 1 2.94 1 .40-6.21 0.005 2.24 0.72-6.97 0.165 0.63 0.23-1.75 0.374 1 .54 0.70-3.40 0.284 0.97 0.95-0.99 0.019

1 .03 0.97-1 .10 0.355 1 4.14 0.39-44.28 0.239 0.28 0.05-1.57 0.144 1 1 .70 0.24-11 .64 0.591 0.71 0.94-1.01 0.665 0.97 0.94-1.01 0.099

1 Not included in multivariate analysis due to non significant value in univariate analysis. IDO expression and CDS+ T-lymplwcytes were included in multivariate analysis irrespective of the p-value.


Univariate Cox regression analyses showed that predominantly well-known factors such as advanced disease, LVSI, 2'.50% myometrial invasion, non-endometrioid tumor type, positive lymph nodes, high tumor grade next to IOOhigh expression and low number of COS+ T-lymphocytes were associated with a worse OSS. Overall, multivariate analysis showed that advanced disease, non-endometrioid tumor type, �50% myometrial invasion, IOOhlgh expression and low number of COS+ T-lymphocytes were independent predictors for a worse OSS. In a subgroup of early stage EC, LVSI, IDOhigh

expression and older age were independent predictors for shorter OSS. For patients with advanced disease, only non-endometrioid tumor type had a negative influence on OSS. Among endometrioid EC, advanced disease, LVSI and low number of COS+

T-lymphocytes were independent predictors for a shorter OSS. In non-endometrioid EC only advanced disease had negative influence on OSS.

Discussion

This study investigated the suppressive mechanism of IOO and evaluated the prognostic influence of 100 expression in a large and well-documented cohort of EC patients. For the first time, the detailed structural localization of IOO in cancer cells was demonstrated by means of electron microscopy. Furthermore, we showed that IDO expression was associated with lower numbers of intratumoral COS+ T-lymphocytes. In addition, IOO expression was independently associated with poor survival in the general cohort of EC patients and among patients with early stage EC. In order to provide a rationale for further research on IOO as new modality for cancer treatment, our results are indicative that mo influences prognosis of EC patients.

mo expression is dependent on IFN-y stimulation, as we showed in several experiments using a human cancer cell line. This finding is in agreement with previous studies showing an IFN-y dependent increase of IDO activity.1,7,29 This finding also demonstrated the correct function of our primary antibody before it was used in further experiments. In order to gain more insight in the localization and distribution of IDO in cancer cells, electron microscopy was used. IDO was localized in the cytoplasm of the cancer cell and predominantly at the level of the cell membrane, which was observed for the first time.

To investigate the mechanism of IDO at a more functional level, concentrations of tryptophan and kynurenine were determined in supernatant and lysates of cancer cells using a validated XLC-MS/MS method.2 We observed that under the influence of IFN-y treatment tryptophan concentrations decreased and kynurenine increased, as measured in supernatant of the cancer cell culture. This demonstrates the functional effect of IOO in the extracellular environment. Here, it is expected that any T-lymphocyte or NK cell in this micro-environment are depleted from the essential amino-acid tryptophan and subsequently more prone to cell cycle arrest and apoptosis.12 However, T-lymphocytes and cancer cells exist in close proximity to each other and one might expect that cancer cells are also depleted from tryptophan. So, why do cancer cells survive and effector T-lymphocytes and NK cells die?

83

Chapter 5

84

Using the same cell cultures, intracellular concentrations of tryptophan and kynurenine were measured. It was observed that, as in the extracellular samples, tryptophan levels decreased while kynurenine concentrations increased under the influence of IFN-y treatment. However, extracellular tryptophan and kynurenine concentrations changed with a factor 339 and 25, respectively, and the intracellular levels of tryptophan and kynurenine only changed with factor 2 in both cases. Possibly, where IDO activity causes a dramatic decrease in tryptophan concentrations in the extracellular environment of cancer cells (median: 0.39 umol/mg protein), intracellular tryptophan concentrations are maintained at an acceptable level for cancer cell survival (median: 1.90 umol/ mg protein). This might be due to the localization of mo which is predominantly at the level of the cell membrane as shown by electron microscopy. This might facilitate the extracellular depletion of tryptophan while maintaining an accurate intracellular level. Another possible explanation might be the upregulation of the system L amino acid transporter 1 (LATl) which transports large amino acids such as tryptophan to the intracellular environment.30 It was previously demonstrated that LATl and other amino acid transporters are upregulated in cancer tissue.31

-33 Furthermore, another study

showed that upregulation of LATl upregulation is related to increased mo activity, functioning as an immune escape mechanism in breast cancer cells.32 This is probably based on the fact that LATl regulates the amino-acid supply and supports the high rate of proliferation in cancer cells. Recently, a novel tryptophan transporter was discovered to be upregulated by IDO in mice and humans.34 Therefore, tryptophan transporters might be involved in maintaining an adequate cellular tryptophan status where the cellular environment of the IDO expressing cells become tryptophan depleted.

Based on our immunohistochemistry results, we showed that IDO expression inversely correlates with the number of CD8+ I-lymphocytes in EC. This finding is in agreement with results found in previous studies in solid tumor showing a reduced number of CD3+ T-lymphocytes7 or CD8+ T-lymphocytes8•19 in tumor tissue with IDOhigh expression compared to IDO10w expression. Next to tryptophan depletion, the suppressive action of IDO on effector I-lymphocytes can also be exerted by the induction and expansion of regulatory T-cells by an !DO-dependent mechanism.16

-18

•35 In contrast to our expecta

tions, we did not find an association between IDO expression and the number of intratumoral Foxp3+ T-lymphocytes, possibly due to the relatively small number of tumor infiltrating Foxp3+ T-lymphocytes (median: 3; IQR: 1-6).

We showed that IDOhigh expression was present in 18.4% of primary EC tissue which is lower compared to the percentage (46%) in the study by Ino et al.10 However, there are some differences in immunohistochemistry analysis between the studies: a TMA containing three cores of representative tumor areas was used in our study, whereas Ino et al. performed IDO immunostaining on whole tissue cores after which "three areas" were evaluated for IDO expression. Unfortunately, the authors did not define if these areas were representative for the entire tumor composition or IDO expression. Furthermore, although IDO expression occurs in a wide range of solid tumors, great differences can exist between tumor types.1

In order to provide a rationale for further research on IDO as new modality for cancer treatment, the prognostic role of IDO expression was determined. IDOhigh expression


independently predicted a poor DSS in the entire EC group and in a subgroup with early stage EC. Based on univariate Cox regression analyses only, IDOhigh expression was associated with DSS in non-endometrioid EC but not in the subgroups of advanced stage EC and an endometrioid tumor type. No association between IDO expression and DFS was observed which is in contrast to results of the previously mentioned study.10

A possible explanation might be the differences in size and composition of the study populations. Where the Japanese study exclusively comprises 80 endometrioid EC, we have used a large cohort of 355 EC patients of all tumor types. Treatment regimens differed slightly between the studies; the Japanese group excluded all patients treated with postoperative radiotherapy. In our population, these patients were not excluded where they represent a significant proportion of the population (58.3%). In general, our study and previous studies in several cancer types have shown that IDO expression has a negative effect on prognosis.6•11 However, we do not agree with the comment of Ino and colleagues that "IDO may be a more reliable prognostic parameter of EC than the currently used clinicopathological factors" where in multivariate analysis, other factors appear to be more important (with a higher Hazard Ratio) than IDO. In addition, more (in vivo) research is required to clarify a possible role of !DO-blocking agents in EC. Our results indicate that possible beneficial effect might only help to improve DSS and but not prevent recurrent disease (DFS).

In summary, the current study investigated the suppressive mechanism of IDO and its prognostic significance in EC patients. We showed that an IFN-y dependent cytoplasmic expression of IDO was exclusively localized just beneath the cell membrane. We showed that !DO-induced tryptophan depletion is predominantly present in the extracellular environment of cancer cells. IDOhigh expression was associated with decreased numbers of CDB+ T-lymphocytes and a decreased DSS. These results might explain why T-lymphocytes are more prone for apoptosis compared to cancer cells whilst in the same microenvironment. New cancer treatment strategies are currently under investigation, including the use of !DO-blocking agents. Based on our results that IDOhigh

expression has a negative influence on DSS in EC patients, this might be a promising new treatment strategy although more research is warranted.

Acknowlegements

The authors would like to thank Baukje-Nynke Hoogeboom, Harry Klip, Jasper Krijnen, Klaas Sjollema, Klaske ten Hoor, Shabnam Jafari, Teus Ruitenbeek and Tineke van der Sluis for their assistance with experiments used for this study.

85

Chapter 5

86

References

1. Uyttenhove C, Pilotte L, Theate I, Stroobant V, Colau D, Parmentier N, et al. Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat Med 2003 10;9:1269-1274.

2. de Jong WH, Smit R, Bakker SJ, de Vries EG, Kema IP. Plasma tryptophan, kynurenine and 3-hydroxykynurenine measurement using automated on-line solid-phase extraction HPLCtandem mass spectrometry. J.Chromatogr.B Analyt.Technol.Biomed.Life Sci. 2009;877:603-609.

3. Munn DH, Zhou M, Attwood JT, Bondarev I, Conway SJ, Marshall B, et al. Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 1998;281:1191-1193.

4. Mellor AL, Munn DH. IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nat.Rev.Immunol. 2004;4:762-774.

5. Munn DH, Sharma MD, Hou D, Baban B, Lee JR, Antonia SJ, et al . Expression of indoleamine 2,3-dioxygenase by plasmacytoid dendritic cells in tumor-draining lymph nodes. J Clin Invest 2004; 114:280-290.

6. Riesenberg R, Weiler C, Spring 0, Eder M, Buchner A, Popp T, et al. Expression of indoleamine 2,3-dioxygenase in tumor endothelial cells correlates with long-term survival of patients with renal cell carcinoma. Clin Cancer Res 2007;13:6993-7002.

7. Brandacher G, Perathoner A, Ladurner R, Schneeberger S, Obrist P, Winkler C, et al . Prognostic value of indoleamine 2,3-dioxygenase expression in colorectal cancer: effect on tumorinfiltrating T cells. Clin Cancer Res 2006;12:1144-1151.

8. Inaba T, Ina K, Kajiyama H, Yamamoto E, Shibata K, Nawa A, et al. Role of the immunosuppressive enzyme indoleamine 2,3-dioxygenase in the progression of ovarian carcinoma. Gynecol Oncol 2009;115:185-192.

9. Inaba T, Ina K, Kajiyama H, Shibata K, Yamamoto E, Kondo S, et al. Indoleamine 2,3-dioxygenase expression predicts impaired survival of invasive cervical cancer patients treated with radical hysterectomy. Gynecol Oncol 2010;117:423-428.

10. Ina K, Yoshida N, Kajiyama H, Shibata K, Yamamoto E, Kidokoro K, et al. Indoleamine 2,3-dioxygenase is a novel prognostic indicator for endometrial cancer. Br J Cancer 2006;95:1555-1561.

11 . Pan K, Wang H, Chen MS, Zhang HK, Weng DS, Zhou J, et al. Expression and prognosis role of indoleamine 2,3-dioxygenase in hepatocellular carcinoma. J Cancer Res Clin Oneal 2008;134:1247-1253.

12. Munn DH, Shafizadeh E, Attwood JT, Bondarev I, Pashine A, Mellor AL. Inhibition of T cell proliferation by macrophage tryptophan catabolism. J Exp Med 1999;189:1363-1372.

13. Della Chiesa M, Carlomagno S, Frumento G, Balsamo M, Cantoni C, Conte R, et al. The tryptophan catabolite L-kynurenine inhibits the surface expression of NKp46- and NKG2D-activating receptors and regulates NK-cell function. Blood 2006;108:4118-4125.

14. Fallarino F, Grohmann U, Vacca C, Bianchi R, Orabona C, Spreca A, et al. T cell apoptosis by tryptophan catabolism. Cell Death Differ 2002;9:1069-1077.

15. Frumento G, Rotondo R, Tonetti M, Damonte G, Benatti U, Ferrara GB. Tryptophan-derived catabolites are responsible for inhibition of T and natural killer cell proliferation induced by indoleamine 2,3-dioxygenase. J Exp Med 2002;196:459-468.

16. Fallarino F, Grohmann U, You S, McGrath BC, Cavener DR, Vacca C, et al. The combined effects of tryptophan starvation and tryptophan catabolites down-regulate T cell receptor zetachain and induce a regulatory phenotype in naive T cells. J Immunol 2006;176:6752-6761 .


17. Baban B, Chandler PR, Sharma MD, Pihkala J, Koni PA, Munn DH, et al. IDO activates regulatory T cells and blocks their conversion into Th17-like T cells. J Irnrnunol 2009;183:2475-2483.

18. Jurgens B, Hainz U, Fuchs D, Felzmann T, Heitger A. Interferon-gamma-triggered indoleamine 2,3-dioxygenase competence in human monocyte-derived dendritic cells induces regulatory activity in allogeneic T cells. Blood 2009;114:3235-3243.

19. Ino K, Yamamoto E, Shibata K, Kajiyama H, Yoshida N, Terauchi M, et al. Inverse correlation between tumoral indoleamine 2,3-dioxygenase expression and tumor-infiltrating lymphocytes in endometrial cancer: its association with disease progression and survival. Clin Cancer Res 2008;14:2310-2317.

20 . Hou DY, Muller AJ, Sharma MD, DuHadaway J, Banerjee T, Johnson M, et al. Inhibition of indolearnine 2,3-dioxygenase in dendritic cells by stereoisomers of 1-methyl-tryptophan correlates with antitumor responses. Cancer Res 2007;67:792-801.

21. Qian F, Villella J, Wallace PK, Mhawech-Fauceglia P, Tario JD,Jr., Andrews C, et al. Efficacy of levo-1-methyl tryptophan and dextro-1-methyl tryptophan in reversing indoleamine-2,3-dioxygenase-mediated arrest of T-cell proliferation in human epithelial ovarian cancer. Cancer Res 2009;69:5498-5504.

22. Amant F, Moerman P, Neven P, Timmerman D, Van Limbergen E, Vergote I. Endometrial cancer. Lancet 2005;366:491-505.

23. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin 2010;60:277-300.

24. de Jong RA, Leffers N, Boezen HM, ten Hoor KA, van der Zee AG, Hollema H, et al. Presence of tumor-infiltrating lymphocytes is an independent prognostic factor in type I and II endometrial cancer. Gynecol Oncol 2009;114:105-110.

25. Bijen CB, de Bock GH, ten Hoor KA, Nijman HW, Hollema H, Maurits MJ. Role of endocervical curettage in the preoperative staging of endometrial carcinoma. Gynecol Oncol 2009;112:521-525.

26. Creutzberg CL, van Putten WL, Koper PC, Lybeert ML, Jobsen JJ, Warlam-Rodenhuis CC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355:1404-1411.

27. Tavassoli FA, Devilee P. World Health Organization: Tumours of the Breast and Female Genital Organs (WHO Classification of Tumours). : IARCPress-WHO; 2003.

28. van den Pol AN, Gores T. Synaptic relationships between neurons containing vasopressin, gastrin-releasing peptide, vasoactive intestinal polypeptide, and glutamate decarboxylase immunoreactivity in the suprachiasmatic nucleus: dual ultrastructural immunocytochernistry with gold-substituted silver peroxidase. J Comp Neurol 1986;252:507-521.

29. Taylor MW, Feng GS. Relationship between interferon-gamma, indoleamine 2,3-dioxygenase, and tryptophan catabolism. FASEB J 1991;5:2516-2522.

30 . Kudo Y, Boyd CA. The role of L-tryptophan transport in L-tryptophan degradation by indoleamine 2,3-dioxygenase in human placental explants . J Physiol 2001;531:417-423 .

31. Fan X, Ross DD, Arakawa H, Ganapathy V, Tamai I, Nakanishi T. Impact of system L amino acid transporter 1 (LATl) on proliferation of human ovarian cancer cells: a possible target for combination therapy with anti-proliferative aminopeptidase inhibitors. Biochem Pharmacol 2010;80:811-818.

32. Travers MT, Gow IF, Barber MC, Thomson J, Sherman DB. Indolearnine 2,3-dioxygenase activity and L-tryptophan transport in human breast cancer cells. Biochim Biophys Acta 2004;1661:106-112.

87

Chapter 5

88

33. Yanagida 0, Kanai Y, Chairoungdua A, Kim DK, Segawa H, Nii T, et al. Human L-type amino acid transporter 1 (LATl): characterization of function and expression in tumor cell lines. Biochim Biophys Acta 2001;1514:291-302.

34. Silk JD, Lakhal S, Laynes R, Vallius L, Karydis I, Marcea C, et al. IDO induces expression of a novel tryptophan transporter in mouse and human tumor cells. J Immunol 2011;187:1617-1625.

35. Chung DJ, Rossi M, Romano E, Ghith J, Yuan J, Munn DH, et al. Indoleamine 2,3-dioxygenaseexpressing mature human monocyte-derived dendritic cells expand potent autologous regulatory T cells. Blood 2009;114:555-563.

Chapter 6

Molecular markers and clinical behavior of

uterine carcinosarcomas;

focus on the epithelial tumor component

Renske A. de Jong

Hans W. Nijman

Tera F. Wijbrandi

Anna K. L. Reyners

H. Marike Boezen

Harry Hollema

Modern Pathology 2011;24:1368-79.

Molecular markers and clinical behavior of uterine carcinosarcomas; focus on the epithelial tumor component

Abstract

Carcinosarcomas (malignant mixed Milllerian tumors) of the uterus are rare and aggressive malignancies consisting of an epithelial (carcinoma) and a mesenchymal (sarcoma) tumor component and are considered as metaplastic endometrial carcinomas. This study evaluated molecular characteristics and clinical behavior of uterine carcinosarcomas to improve treatment regimens in the future. Immunohistochemical expression of estrogen receptor-a and -B, progesterone receptor-A and -B, MLHl, MSH2, MSH6, PTEN (phosphatase and tensin homolog deleted on chromosome 10), p53, B-catenin and cyclin D1 was determined in 40 uterine carcinosarcomas. Immunostaining was compared between epithelial and mesenchymal tumor components. To determine the prognostic role of the epithelial component, clinicopathological data and survival were compared between patients with endometrioid and non-endometrioid epithelial tumor components. To determine prognosis of carcinosarcomas compared to high-risk endometrial carcinomas, clinicopathological characteristics and survival were compared between these patients. Hormone receptor expression occurred infrequently; estrogen receptor-a (8%) and -B (32%), progesterone receptor-A (0%) and -B (23%), next to B-catenin (4%) and cyclin D1 (7%). PTEN, MLHl, MSH2 and MSH6 mutations occurred in 39%, 33%, 22% and 21 %, respectively (based on absent immunostaining). Overexpression of p53 was observed in 38%. Expression patterns of p53, MSH2 and MSH6 corresponded between epithelial and mesenchymal tumor components. In our cohort, the epithelial component caused the majority of metastases (72%) and vascular invasion (70%). Survival tended to be worse for patients with a non-endometrioid epithelial component compared to an endometrioid epithelial component (5-year survival: 26% and 55%, respectively). Survival was worse for patients with uterine carcinosarcomas compared to high-risk endometrial carcinomas (grade 3 endometrioid and non-endometrioid); 5-year survival rates: 42%, 77% and 57%, respectively. Our results support the monoclonal origin of uterine carcinosarcomas. The epithelial component determines prognosis by causing the majority of metastases and vascular invasion. To improve prognosis, treatment should focus on the epithelial tumor component of uterine carcinosarcomas.

91

Chapter 6

92

Introduction

Uterine carcinosarcomas (malignant mixed Miillerian tumors) are rare malignancies of the female genital tract, accounting for 1-5% of uterine malignancies.1 -2 Microscopically, carcinosarcomas consist of two histological malignant components; an epithelial (carcinoma) and a mesenchymal (sarcoma) component. The epithelial component is usually a high grade carcinoma such as papillary serous or clearcell.3 The mesenchymal component may be either homologous or heterologous. The homologous mesenchymal component contains cell types that are normally found in the uterus; stromal sarcoma, fibrosarcoma, undifferentiated sarcoma and leiomyosarcoma. These cell types can either occur as single or mixed tissue type. The heterologous component is composed of other components such as rhabdomyosarcoma, chondrosarcoma, osteosarcoma and liposarcoma.4,5 The proportion of the epithelial and mesenchymal component can vary between individual cases.

Several theories about the origin of the coexistence of two distinctive malignant components in the same tumor have been proposed during the last decades. The 'collision' theory suggests that the epithelial and mesenchymal component originated separately and finally collided in one "mixed" tumor. The 'combination' theory comprises the assumption of a common (epithelial) precursor stem cell with bidirectional differentiation. The 'conversion' theory suggests that the epithelial component is "the driving force": the mesenchymal component is derived from the epithelial component via a metaplastic process.6'7 MoleculJr and immunohistochemical studies suggest that most, but not all, carcinosarcomas are monoclonal, supporting the conversion theory.6,s,9

Therefore, uterine carcinosarcomas are considered as metaplastic endometrial carcinomas. Endometrial carcinomas can be divided in two types based on clinicopathological characteristics; type I consists of endometrioid carcinomas and type II consist of clear-cell and serous papillary carcinomas.7,10 Type I endometrial carcinomas are associated with mutations of DNA mismatch repair (MMR) genes, PTEN (phosphatase and tensin homolog deleted on chromosome 10) mutations, estrogen receptor expression, progesterone receptor expression and aberrant Wnt/ 13-catenin signaling pathway. Type II endometrial carcinomas are characterized by p53 mutations and have a worse prognosis compared to type I endometrial carcinomas. High-risk subtypes of endometrial carcinomas (grade 3 endometrioid and non-endometrioid) show resemblance to the aggressive biological behavior of uterine carcinosarcomas although prognosis of uterine carcinosarcomas is worse.11,12

Uterine carcinosarcomas predominantly occur in postmenopausal women and a higher incidence is found amongst black women compared to white women. 1 Risk factors are similar to endometrial carcinomas; advanced age, obesity, nulliparity and exposure to exogenous estrogen. Furthermore, long-term use of tamoxifen after breast cancer has been associated with development of a uterine carcinosarcoma.4

Stage of disease, myometrial invasion and vascular invasion are important prognostic factors. Due to a high tendency to early extrauterine spread, advanced disease is usually present at time of diagnosis.4,13 Prognosis is poor; 5-year survival rates have been reported between 30% and 45.8% in early stage carcinosarcoma (FIGO stage I/


II) and 0% to 10% in advanced stage carcinosarcoma (FIGO stage III/IV). 11·12

Until recently, uterine carcinosarcomas were considered as a subtype of uterine sarcomas and were treated correspondingly. However, hardly any improvement of prognosis was observed. Based on the fact that uterine carcinosarcomas are currently considered as metaplastic endometrial carcinomas, uterine carcinosarcomas are treated as high-risk endometrial carcinomas. Surgical treatment consists of a total abdominal hysterectomy, bilateral salpingo-oopherectomy, dissection of pelvic and para-aortic lymph nodes and collection of peritoneal cytology. However, responses to present-day adjuvant radiotherapy and chemotherapy are poor and clinical trials are conducted in order to improve prognosis by new treatment strategies. Treatment with adjuvant combined chemotherapy seems the most promising compared to single agent chemotherapy14·15 or radiotherapy.16 However, more research is needed to develop the most effective treatment for patients with uterine carcinosarcomas.

The current study aimed to gain more insight in the molecular characteristics and clinical behavior of carcinosarcomas in order to improve treatment regimens in the future. Therefore, immunohistochemical expression of hormone receptors ( estrogen receptoralpha (ER-a), estrogen receptor-beta (ER-�), progesterone receptor A (PR-A) and progesterone receptor B (PR-B)), MMR proteins (MLHl, MSH2 and MSH6), PTEN, p53, �-catenin and cyclin Dl was determined in a well defined cohort of uterine carcinosarcomas. Expression levels were compared between epithelial and mesenchymal tumor components and between primary and metastatic tumor tissue. In order to determine whether treatment modalities should focus on characteristics of the epithelial tumor component, the prognostic role of the epithelial component was determined. Therefore, clinicopathological data and survival were compared between patients with endometrioid and non-endometrioid epithelial tumor components of uterine carcinosarcomas. Next, clinicopathological characteristics and survival were compared between patients with high-risk endometrial carcinomas and carcinosarcomas in order to determine if carcinosarcomas have a worse prognosis compared to high-risk endometrial carcinomas.

Materials and methods

Patients and treatment Since 1980, tissue samples of patients with gynecological malignancies treated at the Department of Gynecologic Oncology of the University Medical Center Groningen are collected and stored in the tissue storage system of the Pathology Department of the University Medical Center Groningen. Clinicopathological characteristics and followup data of these patients were prospectively collected during standard treatment and were stored in a computerized registration database. For the present study, patients with grade 3, non-endometrioid endometrial carcinoma and carcinosarcomas were selected if diagnosed and treated by a gynecologic oncologist in the University Medical Center Groningen between 1980 and 2006. Of these patients, clinicopathological data were retrieved from hospital and pathology records and compared. For carcinosarcoma patients it was assessed whether sufficient tissue material was available from the pri-

93

Chapter 6

94

mary tumor location and metastatic lesions. Staging occurred after surgical treatment according to the FIGO guidelines.17 Tumors were classified and graded by pathologists according to the World Health Organization (WHO) criteria.18 Follow-up data were completed until February 2010.

Institutional Review Board approval For the present study, all relevant data were retrieved from our computerized database and transferred into a separate, anonymous, password protected database. Patient identity was protected by study-specific, unique patient codes, which were only known to two dedicated data managers, who also have daily responsibility for the larger database. In case of uncertainties with respect to clinicopathological and follow-up data, the larger databases could only be checked through the data managers, thereby ascertaining the protection of patients' identity. Using the registration database all tissue specimens were identified by unique patient numbers and retrieved from the archives of the Department of Pathology. Therefore, according to Dutch law no further Institutional Review Board approval was needed for this study (http: //www.federa.org/).

Tissue microarray construction The tissue microarray method allows simultaneous evaluation of several markers on paraffin embedded tissues from hundreds of tumors.19 For the present study, archival slides of all cases were reviewed and the histopathological classifications of the carcinosarcomas were confirmed by an experienced gynecologic pathologist (HH). Morphologically representative areas of the epithelial and mesenchymal tumor component were marked on haematoxylin & eosin (H&E) stained slides of the paraffin-embedded tissue. Areas of necrosis and areas with severe leukocyte infiltration were avoided. Three core biopsies of 0.6 mm were taken from each tumor component and arrayed on a recipient paraffin block using a tissue microarrayer (Beecher instruments, Silver Spring, Maryland, USA). Adhesion of cores to the recipient block was accomplished by placing the blocks in a 37 °C oven for 15 minutes.

Immunohistochemistry For immunohistochemistry, 4 µm sections were cut from the tissue microarrays and mounted on amino-propyl-ethoxy-silan-coated glass slides (Sigma-Aldrich, Diessenhofen, Germany). Eleven primary antibodies were used for immunohistochemical assessment. Antibodies, antigen retrieval methods and detection techniques are summarized in Table 1. Sections were deparaffinized in xylene and rehydrated in ethanol. Endogenous peroxidase was blocked by incubation in a 0.3 % H

20

2 solution for

30 minutes. Staining was visualized with 3,3' -diaminobenzidine (Vector Laboratories, Burlingame) and slides were counterstained with haematoxylin.

Evaluation of staining Immunohistochemical expression was determined based on intensity and extend of the staining. Intensity was scored as negative (0), weak (1 +), positive (2+) or strong positive (3+ ). Immunostaining for p53 was scored as follows: tumors showing at least 50% positive nuclear expression were considered as having aberrant p53 expression. Positive


Table 1. Antibodies used for immunohistochemical staining.

Antigen Antigen retrieval Primary Company Dilution Detection antibody method

PTEN Citrate (pH 6)1 28H6 Neomarkers3 1:50 Envision

P53 Tris/EDTA (pH 9)1 DO-7 DAKO4 1:1000 Dako En Vision•

B-catenin Citrate (pH 6) 1 Clone 14 BD Transduction 1 :1000 RAMhrp 1:100, Laboratories5 GARhrp 1:100

ER-a Citrate (pH 6)2 6F11 Serotec6 1:20 Goat anti-mouse IgGl/HRP 1:40

ER-� Citrate (pH 6)2 ppg5/10 Serotec6 1:20 RAMbio 1:300, Streptavidine/HRP 1:100

PR-A EDTA (pH 8) 1 hPRa7 Neomarkers3 1:50 RAMP0 1:100, GARpo l:100

PR-B Autoclave2 hPRa2 Neomarkers3 1:50 RAMP0 1:100, GARP0 1:100

Cyclin Dl Tris-HCL1 AM29 Zymed Laboratories7 1:500 RAMP0 1:100, GARP0 1:100

MSH2 G219-1129 Ventana8

MSH6 G70220 BD Transduction 1:400 Laboratories5

MLHl G168-728 Ventana8

1 Sections were boiled in a microwave for 15 minutes. 2 Sections were treated in an autoclave for 3 times 5 minutes at 115°C in blocking reagent (2% block + 0.2% SDS in maleic acid, pH 6.0, Boehringer Mannheim, Mannheim, Germany). 3 Neomarkers, Lab Vision Corporation, Fremont, USA. 4 DAKO, Netherlands BV, Heverlee, Belgium. 5 BD Biosciences, Lexington, KY, USA. 6 Serotec, AbD Serotec, Dusseldo1f, Germany. 7

Zymed, San Fransisco, USA. 8 Ventana, Tucson, USA. 9 The antibody was ready-to-use.

staining of PTEN was defined as presence of > 10% cytoplasmic immunostaining.20 Cydin Dl and hormone receptors expression was considered positive when > 10% tumor cells had moderate to strong nuclear expression. MSH2, MSH6 and MLHl expression was scored as either negative (i.e. total absence of detectable nuclear staining of tumor cells) or positive. [3-catenin was considered positive when at least 10% tumor cells showed nuclear immunohistochemical expression. Two independent researchers (T.W. and H.H.) scored all immunohistochemical stained slides without prior knowledge of clinicopathological data. Discordant cases were reviewed and scores were reassigned on consensus of opinion. Staining was only analyzed when two or more cores were available, each containing more than 20% tumor tissue. In this way, resemblance to whole tissue slides was warranted.

95

Chapter 6

96


percentile) were presented. To establish whether clinicopathological characteristics were associated with expression of MLHl, MSH2, MSH6, PTEN, p53, hormone receptors, �-catenin and cyclin Dl, univariate logistic regression analyses were performed. Expression of the markers was dichotomized according to negative and positive immunostaining and analyzed as dependent variables and clinicopathological characteristics (Table 2) were used as independent variables; odds ratios (ORs) and 95% confidence intervals (95%-CI) were calculated. Associations of immunohistochemical expression (Table 3) between epithelial and mesenchymal tumor component and between primary and metastatic tumor tissue were tested using Chi-square tests (or Fisher 's exact tests, if appropriate). Spearman rank correlation analyses were used to determine correlations between expression of PTEN and MLHl, MSH2 and MSH6. Chi-square tests (or Fisher 's exact tests, if appropriate) or Mann-Whitney U tests were used to assess differences in clinicopathological characteristics (Table 4 and Table 5) between tumor types. Differences in disease specific survival based on immunohistochemical expression or tumor types were plotted using Kaplan-Meier survival curves and evaluated by log-rank tests. Disease specific survival was defined as the time from diagnosis until death due to disease (endometrial carcinoma or uterine carcinosarcoma) or date of last follow-up. All tests were performed two-sided and p-values of <0.05 were considered statistically significant. Analyses were performed using the software package SPSS, version 16.0 for Windows (SPSS Inc. Chicago Illinois, USA).

Results

Patients Between 1980 and 2006, 725 patients were diagnosed and treated for endometrial cancer in the University Medical Center Groningen. Ninety-nine patients (14%) were diagnosed with high-risk endometrial carcinoma; grade 3 endometrioid carcinoma (n=56, 8%), serous papillary carcinoma (n=17, 2%) and clear cell carcinoma (n=26, 4%). Fortythree patients (6%) were diagnosed with carcinosarcoma of the uterus. From a total of 40 carcinosarcoma patients, sufficient paraffin-embedded tumor tissue was available for construction of a tissue microarray. Tumor tissue from the primary tumor location was available of 38 patients and in 32 cases both the epithelial and mesenchymal component could be incorporated on the tissue microarray. Metastatic tumor tissue was available of 18 patients.

Patient characteristics of carcinosarcoma patients Clinicopathological characteristics of carcinosarcoma patients are summarized in Table 2. Median age at time of diagnosis was 66 years (IQR: 59-76). Median time of follow-up was 1.5 years (IQR: 0.8-6.1). Patients were diagnosed with advanced stage of disease (FIGO stage III/IV) in 50% of the cases. In the majority of cases, both tumor components were poorly differentiated (epithelial: 74%; mesenchymal: 92%). Vascular invasion was

Table 2. Clinicopathological characteristics of 40 patients with uterine carcinosarcoma

Characteristics Patients no (n = 40) %1

Stage I 17 42 II 3 8 m 13 33 IV 7 17

Tumor type (epithelial) Endometrioid 25 68 Serous papillary 7 19 Clear cell 2 5 Undifferentiated 3 8 Missing 2 3

Tumor grade (epithelial) Grade 1 1 3 Grade 2 8 23 Grade 3 23 66 Undifferentiated 3 8 Missing 2 5

Tumor type (mesenchymal) Homologous 27 75 Leiomyosarcoma 3 11 Stromal sarcoma 6 22 Not otherwise specified (NOS) 18 67 Heterologous 9 25 Rhabdomyosarcoma 4 45 Chondrosarcoma 1 11 Osteosarcoma 3 33 Stromal sarcoma + chondrosarcoma 1 11 Missing 2 4

Tumor grade (mesenchymal) Low 3 8 High 34 92 Missing 2 3

Myometrial invasion <50% 21 60 >50% 14 40 Missing 3 5

Vascular invasion Negative 12 34 Positive 23 66 By carcinoma 16 70 By sarcoma 6 26 By carcinoma and sarcoma 1 4 Missing 3 5

Recurrent disease No 4 25 62 Yes 15 38 Local 2 13 Pelvic region 3 20 Distant 10 67

1 Percentages exclude missing cases. 2 Missing cases include cases of which no primary tumor tissue was available for analysis (only metastatic tumor tissue present on TMA). When one component accounted for less than 5% of the entire tumor, exact tumor type or grade could not be determined. 3 Missing cases represent patients not treated with TAH-B50. 4 Include patients with residual disease after priman; treatment.

97

Chapter 6

98

present in 23 cases (66%). In 13 out of 18 cases (72%), metastases were caused by the epithelial component only; both components were present in 4 out of 18 cases (22%) and in 1 case (6%) only the mesenchymal component represented metastatic tissue. Total abdominal hysterectomy and bilateral salpingo-oopherectomy was performed in 30% of the patients; more extended surgery with pelvic and/ or para-aortic lymph node dissection was performed in 58%. Adjuvant radiotherapy was given to 19 patients (48%) and adjuvant chemotherapy was given to 5 patients (13%). Due to metastatic disease, 5 patients (13%) were not treated with total abdominal hysterectomy and bilateral salpingo-oopherectomy but received palliative chemotherapy (2 patients) or radiotherapy (3 patients) only. Recurrent disease developed in 15 patients (38%), with a median time to recurrence of 5 months (IQR: 3-11). Three patients (8%) did not have a disease-free interval after surgery. In total, 23 patients (58%) died as a result of disease during our follow-up. The median time between diagnosis and death of disease was 1 year.

Immunohistochemistry Immunohistochemical results are summarized in Table 3.

Hormonal receptors (ER-a, ER-{3, PR-A, PR-B) Positive ER-a expression was observed in 33% of the epithelial component and 17% of the mesenchymal component. Positive immunostaining was present in both components in 8% of the cases. ER-a expression was associated with low tumor grade (OR: 5.4; 95%-CI: 1.1-27.8), an endometrioid epithelial component (OR: 7.4; 95%-CI: 0.8-68.1) and no vascular invasion (OR: 7.2; 95%-CI: 1.5-34.1) (data not shown). ER-a expression was observed in 23% of the metastatic tissue samples. Positive ER-B expression was more frequently observed in the mesenchymal component compared to the epithelial component (56% vs. 37%, respectively). In 8 of 25 cases (32%) positive expression was seen in both components (p=0.041). In metastatic tumor tissue, positive ER-B expression was observed in 67% (epithelial component) and 60% (mesenchymal component). No associations were found between ER-B expression and clinicopathological characteristics. PR-A expression was observed in 5% (epithelial component) and in 10% (mesenchymal component). PR-A expression was not observed in both tumor components simultaneously. No associations were found between PR-A expression and clinicopathological characteristics. PR-B expression was found in 41 % and 35% in the epithelial and the mesenchymal component, respectively. When paired tumor components were present for evaluation, positive PR-B expression was seen in 23% of the tumors. Furthermore, percentages of positive expression were higher in metastatic tissue; 47% and 80% positivity in epithelial and mesenchymal components respectively. PR-B expression was significantly associated with an endometrioid epithelial component (OR: 11.2; 95%-CI 1.2-104.3) and low tumor grade (OR: 4.5; 95%-CI 0.9-22.7) (data not shown).

MMR proteins (MLHl, MSH2 and MSH6) Loss of expression of �1 MMR protein was observed in 12 of 29 carcinosarcomas (41 %). In the epithelial tumor component, absent immunostaining of MLHl, MSH2 and MSH6 was observed in 61 %, 35% and 36%, respectively. Percentages were slightly lower in


the mesenchymal component (MLHl: 40%, MSH2: 29% and MSH6: 27%, respectively). When comparing expression patterns between both tumor components, similarities were observed in the majority of the cases; MLHl (63%, p=0.134), MSH2 (74%, p=0.033) and MSH6 (68%, p=0.050). Tumors showed completely negative immunohistochemical expression for MLHl in 8 of 24 cases (33%), for MSH2 in 6 of 27 cases (22%) and for MSH6 in 6 of 29 cases (21 %). No associations were found between expression of MLHl, MSH2, MSH6 and clinicopathological characteristics.

PTEN Complete loss of PTEN expression was observed in 12 of 31 cases (39%). Expression levels differed slightly between epithelial (64% absent expression) and mesenchymal components (52% absent expression). In the majority of cases, expression was similar in both tumor components (65%, p=0.100). No associations were found between expression of PTEN and clinicopathological characteristics. PTEN expression correlated with MLHl (rs=0.722; p<0.001) and MSH2 expression (r

5=0.440; p=0.010, respectively) in

the epithelial component. Furthermore, absent PTEN expression was more frequently detected in tumors with loss of expression of 2::l MMR protein (7 /12; 58%) compared to tumors without loss of MMR protein expression (4/17, 24%) although this was not significant (p=0.119) (data not shown).

P53 P53 overexpression was observed in 46% and 53% in the epithelial and the mesenchymal component, respectively. In the majority of cases, expression was similar in both tumor components (83%, p<0.001). Furthermore, p53 expression was similar in primary tissue and paired metastatic tissue (p=0.002). Overexpression of p53 was more often observed in a non-endometrioid epithelial tumor component (OR: 16.0; 95%-CI: 1.7-151.1).

{3-catenin Nuclear �-catenin expression was observed in 11 % (epithelial component) and 13% (mesenchymal component). In only one case, nuclear �-catenin expression was observed in both tumor components. �-catenin expression was associated with an endometrioid tumor type (p=0.035) (data not shown). Furthermore, �-catenin expression was similar in primary tissue and paired metastatic tissue (p=0.038).

Cyclin D1 Cyclin D1 expression was observed in 24% (epithelial component) and 23% (mesenchymal component). In 7% of the cases both tumor components showed positive cyclin D1 expression. Similar expression patterns were observed in primary and paired metastatic tumor tissue (p=0.035). No associations were observed between cyclin D1 and clinicopathological parameters. Simultaneous co expression of nuclear �-catenin and cyclin D1 was found in 1 case (3%).

No associations were found between immunohistochemical expression of molecular markers and disease specific survival of uterine carcinosarcoma patients.

99

Table 3. Summary of immunohistochemistry expression in primary and metastatic tumor tissue of uterine carcinosarcomas

Primary tumor tissue Metastatic tumo1· tissue

Marker Entire tumor Epithelial Mese11cl1ymal p-value 4 Epithelial Mesenclzymal p-value 5

N=31 (%)3 compone11t component component compone11t

N = 37 (%)3 N = 33 (%)3 N = 17 (%)3 N = 5 (%)3

ERa No expression 23 (92%) 22 (67%) 25 (83%) 0.656 10 (71 %) 5 (100%) 0.631

Expression 1 2 (8%) 11 (33%) 5 (17%) 4 (23%) 0 (0%)

Missing 6 4 3 3 0

ER{J No expression 17 (68%) 20 (63%) 12 (44%) 0.041 4 (33%) 2 (40%) 0.571

Expression 1 8 (32%) 12 (37%) 15 (56%) 8 (67%) 3 (60%)

Missing 6 5 6 5 0

PR-A No expression 30 (100%) 35 (95%) 27 (90%) 0.735 14 (93%) 5 (100%) n.a.

Expression 1 0 (0%) 2 (5%) 3 (10%) 1 (7%) 0 (0%)

Missing 1 3 2 0

PR-B No expression 23 (77%) 22 (59%) 20 (65%) 0.132 8 (53%) 1 (20%) 0.293

Expression 1 7 (23%) 15 (41 %) 11 (35%) 7 (47%) 4 (80%)

Missing 1 0 2 2 0

MLHl No expression2 8 (33%) 19 (61 %) 11 (40%) 0.134 1 (8%) 1 (25%) 0.377

Expression 16 (67%) 12 (39%) 16 (59%) 12 (92%) 3 (75%)

Missing 7 6 6 4 1

MSH2 No expression2 6 (22%) 12 (35%) 9 (29%) 0.033 3 (21 %) 1 (20%) 0.522

Expression 21 (78%) 22 (65%) 22 (71 %) 11 (79%) 4 (80%)

Missing 4 3 2 3 0

MSH6 No expression2 6 (21 %) 13 (36%) 8 (27%) 0.050 4 (31 %) 0 (0%) 0.317

Expression 22 (79%) 23 (64%) 22 (73%) 9 (69%) 5 (100%)

Missing 3 1 4 4 0

PTEN No expression2 12 (39%) 23 (64%) 16 (52%) 0.100 8 (53%) 4 (100%) 0.835

Expression 19 (61 %) 13 (36%) 15 (48%) 7 (47%) 0 (0%)

Missing 0 1 2 2 1

P53 No expression 18 (62%) 20 (54%) 14 (47%) <0.001 6 (46%) 1 (25%) 0.002

Expression 1 11 (38%) 17 (46%) 16 (53%) 7 (54%) 3 (75%)

Missing 2 0 3 4 1

{3-cateniu No expression 27 (96%) 31 (89%) 26 (87%) 0.270 10 (77%) 4 (100% 0.038

Expression 1 1 (4%) 4 (11 %) 4 (13%) 3 (23%) 0 (0%)

Missing 3 2 3 4 1

Cyclin D1 No expression 28 (93%) 28 (76%) 24 (77%) 0.565 8 (62%) 4 (80%) 0.035

Expression 1 2 (7%) 9 (24%) 7 (23 %) 5 (38%) 1 (20%)

Missing 1 0 2 4 0

1 Expression was considered positive when both components showed positive immunostaining. 2 Expression was considered negative when both components showed

absent immunostaining. 3 Percentages exclude missing cases. 4 Chi-square test comparing expression between epithelial and mesenchymal component in primary tumor

tissue (metastatic was not compared due to small sample size). 5 Chi-square test comparing expression between primary and metastatic tumor tissue.

N.a. = not applicable

Chapter 6

102

Comparison of endometrioid and non-endometrioid epithelial tumor components of carcinosarcomas In the majority of carcinosarcoma patients, the epithelial tumor component accounted for metastases (72%) and vascular invasion (70%). We investigated whether clinicopathological characteristics and disease specific survival of patients differed between an endometrioid and a non-endometrioid epithelial tumor component. For this analysis, undifferentiated tumor type (n=3) or unknown tumor type (n=3) in the epithelial component were excluded. As shown in Table 4, frequencies of positive peritoneal washings (tumor cells present), omental metastases and recurrent disease were not significantly different between endometrioid and non-endometrioid epithelial components of carcinosarcomas. However, peritoneal metastases tended to occur more fre-

Table 4. Characteristics according to histological subtype of the epithelial tumor component of uterine carcinosarcomas

Characteristics All cases E11domet,-ioid Non-endometrioid p-value 3

(11=40) 1 (%) z (11=25) (%) 2 (n=9) (%) 2

Age 0.1834

Median (IQR) 66 (59-76) 66 (59-75) 67 (63-84) Peritoneal washing 0.518

Negative 15 (71 %) 9 (75%) 4 (45%) Positive 4 (19%) 1 (8%) 2 (22%) Suspect 2 (10%) 2 (17%) 3 (33%) Missing 5 19 13 0

Omental metastases 0.156 No 25 (81 %) 19 (95%) 5 (71 %) Yes 6 (19%) 1 (5%) 2 (29%) Missing 9 5 2

Peritoneal metastases 0.063 No 30 (83%) 21 (95%) 6 (67%) Yes 6 (17%) 1 (5%) 3 (33%) Missing 4 3 0

Recurrent disease 0.224 No 25 (63%) 18 (72%) 4 (44%) Yes 15 (38%) 7 (18%) 5 (56%) Local 2 (13%) 1 (14%) 1 (20%) Pelvic region 3 (20%) 2 (29%) 1 (20%) Distant 10 (67%) 4 (57%) 3 (60%)

Follow-up 0.083 Alive 9 (22%) 9 (36%) 0 (0%) Death due to disease 23 (58%) 11 (44%) 6 (67%) Death due to other disease 8 (20%) 5 (20%) 3 (33%) Missing 0 0 0

1An epithelial component consisting of undifferentiated tumor type (n=3) or unknown tumor type (n=3) were excluded for further analysis. 2 Percentages exclude missing cases. 3 Chi-square test used, character-istics divided into two categories. 4 Mann-Whitney U-test used. 5 Missing cases: patients without primary surgery (n=S) or material not enough for definitive conclusion.


quently in patients with a non-endometrioid epithelial tumor component (p=0.063). During follow-up, 9 patients were still alive without evidence of disease. All these patients were diagnosed with an endometrioid epithelial tumor component (p=0.083). Patients with a non-endometrioid epithelial component tended to have a worse disease specific survival (5-year survival: 26%) compared to patients with an endometrioid epithelial component (5-year survival: 55%) (p=0.104) (Figure 1).

1 0

0.8

0.6

0 4

0 2

0.0

1 0

0 8

O G

0 4

0 2

0.0

End1111elri>ld

.- - - - - - - - - - - - -. :_ Non-endometnoi:I :

tO

follow.up (years)

,��+: I •: I \

I <-;.

... ,'+

··;···:,.._. : ........ ....,···: ... . .

I Orade 3 endllTielriDd carc111111a I : .� ...... ,

· ....... . --···"T······•-t-··-t-••···+···· . · . .

i Non-endcmelrlold carcirona !

... T - -H - -r - - -- -

,_.,. ---+ - -;- - - - - - -

follow.up (years)

, Uterine carcinosa,coma • •- - - - - - - - - - - - - - - - - - - - I

10

Figure 1. Disease specific survival according to tumor type in the epithelial component of uterine carcinosarcoma (endometrioid vs. nonendometrioid) (p== 0.104, Log-rank)

Figure 2. Disease specific survival according to tumor type (grade 3 endometrioid carcinoma, non-endometrioid carcinoma and uterine carcinosarcoma) (p< 0.001, Logrank)

103

Chapter 6

104

Comparison of characteristics between carcinosarcomas and high-risk endometrial carcinomas Clinicopathological characteristics and disease specific survival were compared between high-risk endometrial carcinomas (n=99) and uterine carcinosarcomas (n=40) (Table 5). Grade 3 endometrioid endometrial carcinomas more frequently showed >50% myometrial invasion compared to uterine carcinosarcomas (p=0.009). Other clinicopathological characteristics did not differ between these subtypes. Patients with uterine carcinosarcomas had a worse disease specific survival (5-year survival: 42%) compared to non-endometrioid carcinoma (5-year survival: 57%) and grade 3 endometrioid carcinoma (5-year survival: 77%) (p<0.001) (figure 2).

Table 5. Comparison of characteristics between high-risk endometrial carcinomas and uterine carcinosar-comas

Grade 3 Non-Characteristics endometrioid endometrioid Carcinosarcoma p-value

(n=56) (%)1 (n=43) (%)1 (n=40) (%)1

Age n.s. 2,3

Median (I QR) 61 (55-72) 67 (59 74) 66 (59-76) Stage n.s. 3A

I 18 (32%) 14 (33%) 17 (42%) II 11 (20%) 4 (9%) 3 (8%) III 20 (36%) 15 (35%) 13 (33%) IV 7 (12%) 10 (23%) 7 (17%)

Myometrial invasion 0.009 4,5

<50% 18 (32%) 22 (54%) 21 (60%) >50% 38 (68%) 19 (46%) 14 (40%) Missing 0 2 5

Vascular invasion 0.072 4•6

Negative 25 (45%) 20 (56%) 1 2 (34%) Positive 30 (55%) 16 (44%) 23 (66%) Missing 1 7 5

Peritoneal washing n.s. 3A Negative 40 (80%) 23 (70%) 15 (71 %) Positive 10 (20%) 10 (30%) 4 (19%) Missing 6 10 21

Recurrent disease n.s. 3.4

No 39 (70%) 32 (74%) 25 (63%) Yes 17 (30%) 11 (26%) 15 (38%) Local 3 (18%) 3 (27%) 2 (13%) Pelvic region 2 (11 %) 1 (9%) 3 (20%) Distant 12 (71 %) 7 (64%) 10 (67%

Follow-up <0.001 3•4

Death due to disease 11/56 (20%) 16/43 (37%) 23/40 (58%)

1 Percentages exclude missing cases. 2 Mann-Whitney U-test used. 3 Carcinosarcoma compared to grade 3 endometrioid and non-endometrioid carcinoma. 4 Chi-square test used, characteristics divided into two categories. 5 Carcinosarcoma compared to grade 3 endometrioid carcinoma. 6 Carcinosarcoma compared to non-endometrioid carcinoma. n.s.= not significant.


Discussion

The present study investigated imrnunohistochemical expression of 11 markers in a welldefined cohort of patients with carcinosarcomas of the uterus, all treated at the University Medical Center Groningen. Immunohistochemical expression was compared between the epithelial and mesenchymal tumor components of uterine carcinosarcomas.

Overexpression of p53 (a tumor suppressor gene, located on chromosome 17ql3.l) showed a high concordance between both tumor components. In addition, p53 overexpression highly correlated between primary and metastatic tumor tissue. Next to p53, mutations in mismatch repair (MMR) genes are an early event in tumorigenesis.21 It has been shown that loss of immunohistochemical staining of MMR proteins (MLHl, MSH2 and MSH6) correlates to the corresponding MMR gene mutation.22 We observed that expression levels of MSH2 and MSH6 correlated between epithelial and mesenchymal tumor components. Above-mentioned results are in line with other studies and confirm the monoclonal origin of uterine carcinosarcomas.8•9·23-25

The estrogen receptor (ER) and progesterone receptor (PR) are important in growth, differentiation and function of reproductive tissues. In the last decade, two subtypes of ER (ER-a and ER-f3) and PR (PR-A and PR-B) have been discovered. It's importance in tumor development and prognosis has been studied in endometrial carcinomas but reports in uterine carcinosarcomas are scarce.26-29 In the current study, ER-a expression was associated with low tumor grade, an endometrioid epithelial tumor component and no vascular invasion, which is in agreement with results in endometrial carcinomas.30 Furthermore, we observed that ER-a was mainly expressed in the epithelial component in contrast to ER-f3 which was more predominantly expressed in the mesenchymal component. Overall, ER-f3 was more frequently expressed than ER-a in our population. These results are in line with two previous reports.26·27 To our knowledge, we are the first to determine subtype expression of PR (PR-A and PR-B) in uterine carcinosarcomas. PR-A was less frequently expressed than PR-B (5% and 10% vs. 41 % and 35% in epithelial and mesenchymal components, respectively). Expression of PR-B was associated with an endometrioid epithelial tumor component, similar to results in endometrial carcinomas.31 Two developmental pathways can be distinguished in endometrial carcinomas; type I endometrial cancers arise on background of hyperplasia after unopposed estrogen stimulation, type II endometrial cancers are not estrogen driven. 10

The fact that ER-a and PR-B are associated with low grade and an endometrioid tumor type in our population suggest a similar pathway in uterine carcinosarcomas. Probably, tumors with an endometrioid epithelial tumor type develop under the influence of estrogen and tumors with a non-endometrioid tumor type develop independent of estrogen. During dedifferentiation to the sarcomatous component, loss of ER-a occurs whereas ER-f3 is more frequently expressed during progression of disease, a mechanism which has been shown previously in endometrial carcinomas.30

In type I endometrial carcinoma, microsatellite instability is a frequent phenomenon with incidences ranging from 20-90% compared to 0-11 % in type II endometrial carcinoma.2,32,33 Microsatellite instability is caused by an inability of the mismatch repair

105

Chapter 6

106

system to cut out and replace the mismatching DNA strains due to methylation or mutation of its proteins (MLHl, MSH2 and MSH6).34 Two previous studies demonstrated that microsatellite instability was present in 5%35 and 23.3%21 of uterine carcinosarcomas. The latter study observed that microsatellite instability is mainly a feature of the epithelial component21 This is in agreement with our finding; loss of MMR protein expression was more frequently observed in the epithelial component compared to the mesenchymal component, although expression levels corresponded between both tumor components in the majority of the cases. Furthermore, we observed that tumors with loss of expression of 2':l MMR protein more frequently had absent PTEN immunostaining which is in agreement with previous reports in endometrial carcinomas.36·37

PTEN (phosphatase and tensin homolog deleted on chromosome 10) acts as a tumor suppressor gene trough the action of its phosphatase protein product. PTEN mutations more frequently occur in type I endometrial carcinomas (35-55%) compared to type II endometrial carcinomas (5-11 %).2·37·38 Mutation or dysfunction of PTEN can be seen as negative immunohistochemistry staining which was the case in 39% of our study population. One previous study reported PTEN mutations in 14.3% of uterine carcinosarcomas.38 A possible explanation for these different percentages is that the latter study was performed in a smaller study population and different detection methods for PTEN mutations were used.

Previous reports have shown that p53 overexpression was present in 28% to 84% of carcinosarcomas8·9•23·25•39 compared to 38% in our population. Mutant or altered p53 gene protein has a prolonged half-life and accumulates to detectable immunohistochemical levels. Overexpression of p53 is typically present in 90% of non-endometrioid endometrial carcinomas compared to 10% in endometrioid endometrial carcinomas.3 In our cohort, overexpression was associated with a non-endometrioid epithelial tumor component of uterine carcinosarcomas.

Aberrant activation of the Wnt signaling pathway plays an important role in the tumorigenesis of a wide range of tumors40 in which r3-catenin has a crucial role. Mutations of r3-catenin result in stabilization of a protein that resists degradation, leading to nuclear accumulation which can be demonstrated by immunohistochemistry.41 Mutations of r3-catenin are considered an early event in tumorigenesis.2 The reported frequency of r3-catenin mutations in type I endometrial carcinomas ranges from 14-44% compared to 0-5% in type II endometrial carcinomas.3 To date, only two studies reported on this subject in uterine carcinosarcomas.24·41 In order to determine the function of the aberrant Wnt signaling in carcinosarcomas of the uterus, we determined r3-catenin and cyclin D1 expression (which is a direct target gene of r3-catenin). In our population, nuclear r3-catenin expression was present in 11 % and 13% in the epithelial and mesenchymal component of carcinosarcomas, respectively. Percentages resemble results found in type I endometrial carcinoma.3 In only one tumor, simultaneous expression of nuclear r3-catenin in both tumor components was observed. Expression differences between tumor components was reported previously, but exact percentages are lacking.24 Previously, positive nuclear r3-catenin immunostaining was detected in 86% of uterine carcinosarcomas41 which is higher compared to our result. However, this study population was smaller (n=7) and different techniques for immunohistochemistry analysis were


used. To our knowledge, we are the first to determine cyclin Dl expression in uterine carcinosarcomas. Positive expression was found in 24% and 23% (epithelial and mesenchymal component, respectively). Co-expression of nuclear �-catenin and cyclin Dl was an infrequent observation; 3% of epithelial components and 3% of mesenchymal components. These results suggest that an activated Wnt/�-catenin pathway is a rare event in uterine carcinosarcomas.

We and others showed that uterine carcinosarcomas have a more aggressive biological behavior compared to high-risk endometrial carcinomas, resulting in a worse disease specific survival. 11 -12 Evidence is emerging that the epithelial component is the "driving force" in this tumor type.5A2 In our population, we observed that the epithelial component was responsible for the majority of metastases (72%) and vascular invasion (70%). In addition, patients with a non-endometrioid epithelial tumor component tended to have more peritoneal metastases compared to patients with an endometrioid tumor component. Next, patients with non-endometrioid epithelial component tended to have a worse disease specific survival compared to patients with an endometrioid epithelial component. Although not statistically significant, these findings suggest that the developmental pathway of uterine carcinosarcomas is similar to endometrial carcinomas and can be divided in type I and type II (according to the epithelial tumor component).

In summary, the current study investigated molecular markers and clinical characteristics of uterine carcinosarcomas. Immunohistochemistry expression of p53, MSH2 and MSH6 highly corresponded between epithelial and mesenchymal components confirming the monoclonal origin of uterine carcinosarcomas. Furthermore, immunohistochemistry results showed similarities to endometrial carcinomas: p53 expression was associated with a non-endometrioid epithelial tumor component and expression patterns of MMR proteins, PTEN and hormone receptors resembled results previously found in type I endometrial carcinomas. In our population, the epithelial component caused the majority of metastases and vascular invasion. Above-mentioned results show that uterine carcinosarcomas are metaplastic endometrial carcinomas with similar developmental pathways. Currently, uterine carcinosarcomas are treated as highrisk endometrial carcinoma, which is justified based on these results. However, patients with uterine carcinosarcomas have a worse disease specific survival compared to highrisk endometrial carcinomas. Therefore, future research is needed to improve therapy and should focus on characteristics of the epithelial component of carcinosarcomas.

Acknowledgements

The authors thank Mrs. Tineke van der Sluis for her assistance with the construction of the tissue microarrays and immunohistochemistry.

107

Chapter 6

108

References


2. Lax SF. Molecular genetic changes in epithelial, strornal and mixed neoplasms of the endornetriurn. Pathology 2007;39:46-54.

3. Sarnarnthai N, Hall K, Yeh IT. Molecular profiling of endornetrial malignancies. Obstet Gynecol Int 2010;2010:162363.

4. McCluggage WG, Abdulkader M, Price JH, et al. Uterine carcinosarcornas in patients receiving tarnoxifen. A report of 19 cases. Int J Gynecol Cancer 2000;10: 280-4.

5. Silverberg SG, Major FJ, Blessing JA, et al. Carcinosarcorna (malignant mixed rnesoderrnal tumor) of the uterus. A Gynecologic Oncology Group pathologic study of 203 cases. Int J Gynecol Pathol 1990;9:l-19.

6. McCluggage WG. Uterine carcinosarcornas (malignant mixed Mullerian tumors) are metaplastic carcinomas. Int J Gynecol Cancer 2002;12:687-90.

7. McCluggage WG, Robboy SJ. Mesenchyrnal uterine tumors and adenornyosis. In: Robboy SJ, Mutter GL, Prat J, Bentley RC, Russell P, Anderson MC, editors. Pathology of the female reproductive tract. 2edn. Churchill Livingstone Elsevier; 2009. pp 427-56.

8. Kounelis S, Jones MW, Papadaki H, et al. Carcinosarcornas (malignant mixed rnullerian tumors) of the female genital tract: comparative molecular analysis of epithelial and rnesenchyrnal components. Hurn Pathol 1998;29:82-7.

9. Wada H, Enomoto T, Fujita M, et al. Molecular evidence that most but not all carcinosarcornas of the uterus are combination tumors. Cancer Res 1997;57:5379-85.

10. Bokhrnan JV. Two pathogenetic types of endornetrial carcinoma. Gynecol Oncol 1983;15:10-7.

11. Arnant F, Cadron I, Fuso L, et al. Endornetrial carcinosarcornas have a different prognosis and pattern of spread compared to high-risk epithelial endornetrial cancer. Gynecol Oneal 2005;98:274-80.

12. George E, Lillernoe TJ, Twiggs LB, Perrone T. Malignant mixed rnullerian tumor versus highgrade endornetrial carcinoma and aggressive variants of endornetrial carcinoma: a comparative analysis of survival. Int J Gynecol Pathol 1995;14:39-44.

13. D' Angelo E, Prat J. Uterine sarcomas: A review. Gynecol Oncol 2010;116:131-9.

14. Homesley HD, Filiaci V, Markman M, et al. Phase III trial of ifosfarnide with or without paclitaxel in advanced uterine carcinosarcorna: a Gynecologic Oncology Group Study. J Clin Oneal 2007;25:526-31.

15. Powell MA, Filiaci VL, Rose PG, et al. Phase II evaluation of paclitaxel and carboplatin in the treatment of carcinosarcorna of the uterus: a Gynecologic Oncology Group study. J Clin Oneal 2010;28:2727-31.

16. Wolfson AH, Brady MF, Rocereto T, et al. A gynecologic oncology group randomized phase III trial of whole abdominal irradiation (WAI) vs. cisplatin-ifosfarnide and rnesna (CIM) as postsurgical therapy in stage I-IV carcinosarcorna (CS) of the uterus. Gynecol Oncol 2007;107:177-85.

17. Cancer Cornittee of the International Federation of Gynaecology and Obstetrics (1986) . Gynecol Oneal 2008;25:383-5.


18. Silverberg SG, Kurman RJ, Nogales F et al. Tumours of the uterine corpus, In: Tavassoli FA, Devilee P (eds). Pathology and Genetics of Tumours of the Breast and Female Genital Organs (World Health Organization Classification of Tumours). 1st edn. IARCPress: Lyon; 2003. pp 217-57.

19. Kononen J, Bubendorf L, Kallioniemi A, et al. Tissue microarrays for high-throughput molecular profiling of tumor specimens. Nat Med 1998;4:844-7.

20. Noordhuis MG, Eijsink JJ, ten Hoor KA, et al. Expression of epidermal growth factor receptor (EGFR) and activated EGFR predict poor response to (chemo)radiation and survival in cervical cancer. Clin Cancer Res 2009;15:7389-97.

21. Taylor NP, Zighelboim I, Buettner PC, et al. DNA mismatch repair and TP53 defects are early events in uterine carcinosarcoma tumorigenesis. Mod Pathol 2006;19:1333-8.

22. Niessen RC, Berends MJ, Wu Y, et al. Identification of mismatch repair gene mutations in young patients with colorectal cancer and in patients with multiple tumours associated with hereditary non-polyposis colorectal cancer. Gut 2006;55:1781-8.

23. Abargel A, Avinoach I, Kravtsov V, et al. Expression of p27 and p53: comparative analysis of uterine carcinosarcoma and endometrial carcinoma. Int J Gynecol Cancer 2004;14:354-9.

24. Saegusa M, Hashimura M, Kuwata T, Okayasu I. Requirement of the Akt/beta-catenin pathway for uterine carcinosarcoma genesis, modulating E-cadherin expression through the transactivation of slug. Am J Pathol 2009;174:2107-15.

25. Sherman ME, Bur ME, Kurman RJ. p53 in endometrial cancer and its putative precursors: evidence for diverse pathways of tumorigenesis. Hum Pathol 1995;26:1268-74.

26. Ansink AC, Cross PA, Scorer P, et al. The hormonal receptor status of uterine carcinosarcomas (mixed mullerian tumours): an immunohistochemical study. J Clin Pathol 1997;50:328-31.

27. Huang GS, Arend RC, Li M, et al. Tissue microarray analysis of hormonal signaling pathways in uterine carcinosarcoma. Am J Obstet Gynecol 2009;200:457-5.

28. Ioffe YJ, Li AJ, Walsh CS, et al. Hormone receptor expression in uterine sarcomas: prognostic and therapeutic roles. Gynecol Oncol 2009;115:466-71.

29. Jazaeri AA, Nunes KJ, Dalton MS, et al. Well-differentiated endometrial adenocarcinomas and poorly differentiated mixed mullerian tumors have altered ER and PR isoform expression. Oncogene 2001;20:6965-9.

30. Jongen V H, Briet JM, de Jong RA, et al. Aromatase, cyclooxygenase 2, HER-2/neu, and p53 as prognostic factors in endometrioid endometrial cancer. Int J Gynecol Cancer 2009;19:670-6.

31. Shabani N, Kuhn C, Kunze S, et al. Prognostic significance of oestrogen receptor alpha (ERalpha) and beta (ERbeta), progesterone receptor A (PR-A) and B (PR-B) in endometrial carcinomas. Eur J Cancer 2007;43:2434-44.

32. Catasus L, Machin P, Matias-Guiu X, Prat J. Microsatellite instability in endometrial carcinomas: clinicopathologic correlations in a series of 42 cases. Hum Pathol 1998;29:1160-4.

33. Karamurzin Y, Rutgers JK. DNA mismatch repair deficiency in endometrial carcinoma. Int J Gynecol Pathol 2009;28: 239-55.

34. Arabi H, Guan H, Kumar S, et al. Impact of microsatellite instability (MSI) on survival in high grade endometrial carcinoma. Gynecol Oncol 2009;113:153-8.

35. Amant F, Dorfling CM, Dreyer L, et al. Microsatellite instability in uterine sarcomas. Int J Gynecol Cancer 2001;11:218-23.

36. Bilbao C, Rodriguez G, Ramirez R, et al. The relationship between microsatellite instability and PTEN gene mutations in endometrial cancer. Int J Cancer 2006;119:563-70.

109

Chapter 6

110

37. Risinger JI , Hayes K, Maxwell GL, et al. PTEN mutation in endometrial cancers is associated with favorable clinical and pathologic characteristics. Clin Cancer Res 1998;4:3005-10.

38. Amant F, de la Rey M, Dorfling CM, et al. PT EN mutations in uterine sarcomas. Gynecol Oneal

2002;85:165-9.

39. Lee SJ, Kim HS, Kim HS, et al. I mmunohistochemical study of DNA topoisomerase I , p53, and Ki-67 in uterine carcinosarcomas. Hum Pathol 2007;38:1226-31.

40. Kurihara S, Oda Y, Ohishi Y, et al. Coincident expression of beta-catenin and cyclin Dl in endometrial stromal tumors and related high-grade sarcomas. Mod Pathol 2010;23: 225-34.

41. Ng T L, Gown AM, Barry T S, et al. Nuclear beta-catenin in mesenchymal tumors. Mod Pathol 2005;18:68-74.

42. Sreenan JJ, Hart WR. Carcinosarcomas of the female genital tract. A pathologic study of 29 metastatic tumors: further evidence for the dominant role of the epithelial component and the conversion theory of histogenesis. Am J Surg Pathol 1995;19:666-74.

Chapter 7

Less gastrointestinal toxicity after adjuvant

radiotherapy on a small pelvic field compared to

a standard pelvic field in patients with


Renske A de Jong

Elisabeth Pras

H. Marike Boezen

Ate G .J. van der Zee

Marian J.E. Maurits

Henriette J. G. Arts

Jan G. Aalders

Annerie Slot

Paul R. Timmer

Harry Hollema

Hans W. Nijman

International Journal of Gynecological Cancer

2012;22:1177-86.

Less gastrointestinal toxicity after adjuvant radiotherapy on a small pelvic field compared to a standard pelvic field in patients with endometrial carcinoma

Abstract

Objective Radiotherapy is associated with short and long-term morbidity. This study compared toxicity rates among endometrial carcinoma (EC) patients treated with adjuvant external beam radiation therapy (EBRT) on a small pelvic field (SmPF) in comparison with a standard pelvic field (StPF) or an extended field (EF).

Methods EC patients preoperatively diagnosed with high-grade histology (grade 3 endometrioid, serous-papillary, clear cell and mixed tumor type) or cervical involvement were treated with total abdominal hysterectomy, bilateral salpingo-oopherectomy and lymphadenectomy in the University Medical Center Groningen between 1999 and 2008. Patients who received adjuvant EBRT were included in this study. EBRT on SmPF (includes only the central pelvis and proximal vagina) was applied in case of negative lymph nodes after adequate lymphadenectomy (�10 lymph nodes removed at bilateral obturator and external iliac nodal stations). In case of positive pelvic lymph nodes or inadequate lymphadenectomy, EBRT on StPF was given. EBRT on EF was applied in case of common iliac and/ or para-aortic lymph node metastases. Retrospectively, acute toxicity was scored during radiotherapy whereas late toxicity was scored from 3 months onwards post treatment by the Common Terminology Criteria for Adverse Events v3.0.

Results Toxicity could be evaluated in 75 patients treated with SmPF (n=33), StPF (n=28) and EF EBRT (n=14). The majority of patients with late adverse events had also reported toxicity during radiotherapy (71 %). The most common late adverse events were gastrointestinal tract related, more frequently present the StPF group (60.7%) compared to SmPF (33.3%) (p=0.032). Especially nausea and anorexia were more frequent in the StPF group (32.1 %) compared to the SmPF group (3.0%) (p=0.004), as well as ileus (14.3% vs. 0%, p=0.039, respectively).

Conclusion Treatment with adjuvant EBRT on a SmPF results in less gastrointestinal late side effects than EBRT on a StPF in surgically staged EC patients.

113

Chapter 7

114

Introduction

Radiotherapy can be associated with short and long-term morbidity. In patients with endometrial carcinoma (EC), adjuvant radiotherapy is used to reduce the rate of locoregional recurrences, but without a beneficial effect on survival.1-4 Factors related to risk of radiotherapy-related toxicity are treatment volume, daily dose, total dose, radiotherapy technique and age.5•8 The most predominant side effects after EBRT occur in the genitourinary and the gastrointestinal tract; particularly small bowel complications.7·8 The PORTEC-2 study showed that patients with high-intermediate risk EC can be treated safely with vaginal brachytherapy (VBT) alone instead of external beam radiation therapy (EBRT).5•9 Therefore, EBRT is currently only indicated for patients with high-risk EC or advanced stage of disease (FIGO III). Furthermore, toxicity due to EBRT is higher after previous lymphadenectomy.6

The routine use of lymphadenectomy in addition to the standard hysterectomy and bilateral salpingo-oophorectomy is still subject of debate. Results of two randomized controlled trials showed that pelvic lymphadenectomy in early stage EC has no therapeutic benefit10•1 1, supporting the general idea that lymphadenectomy can be omitted in early stage, low-risk EC.12·13 In contrast, several retrospective studies demonstrated that survival of high-intermediate risk EC patients improved with the extent of lymphadenectomy.14-17 Especially patients with high-risk EC have increased risk for lymph node metastases, distant metastases and death due to EC. In these patients, adequate surgical staging might be helpful to determine prognosis and adjuvant treatment strategies.17·18

In accordance with a study in stage 1B cervical carcinoma patients 19, patients with highrisk, early stage EC without proven lymph node metastases have an indication for adjuvant EBRT on a small pelvic field (SmPF) (including the central pelvis only) in order to reduce treatment-related morbidities without compromising control of central pelvic recurrences. Patients with positive lymph nodes or inadequate lymphadenectomy are treated with EBRT on a standard pelvic field (StPF). Extended field (EF) EBRT is given in case of proven common iliac or para-aortic lymph node metastases.

This study compared toxicity rates among patients treated with adjuvant EBRT on SmPF with those treated with StPF or EF depending on the lymph node status. In this way, the effect of the radiation volume on the occurrence of adverse events was evaluated.

Materials and methods

Patients and treatment Patients with EC were consecutively selected when preoperatively diagnosed between 1999 and 2008 in the University Medical Center Groningen (UMCG) with high-grade histology (grade 3 endometrioid, serous-papillary, clear cell and mixed tumor type) or cervical involvement. These patients were indicated for surgical treatment with to-


tal abdominal hysterectomy and bilateral salpingo-oophorectomy (TAH-BSO), pelvic and para-aortic lymphadenectomy. Lymphadenectomy was considered 'adequate' when at least 10 lymph nodes at the bilateral obturator and external iliac nodal stations had been removed based on the pathology report.20 When fewer than 10 lymph nodes had been removed or one of the nodal stations (obturator and external iliac) had been omitted, lymphadenectomy was considered 'inadequate'. Usually, lymphadenectomy also included lymph node dissection at the common iliac and para-aortic nodal stations unless contra-indications (high age, significant comorbidity of the patient or unexpected situations during surgery) were present. Staging occurred according to the FIGO guidelines. Tumors were classified and graded according to the World Health Organization (WHO) criteria. Patients not treated with primary surgery, those who had a secondary malignancy at time of diagnosis or treated with preoperative radiotherapy were excluded. EBRT was given when 2 out of 3 risk factors (i.e. grade 3 histology, age over 60 years or deep myometrial invasion) were present.1•9 Patients with clear cell or serous papillary tumor type or cervical involvement were treated with EBRT irrespective of other risk factors. Radiotherapy was started within 4-8 weeks after surgery in one of the three radiation oncology centers in the Northern part of the Netherlands (Isala Clinics, Zwolle; Radiotherapy Institute Friesland, Leeuwarden; UMCG, Groningen). Patients without lymph node metastases after adequate lymphadenectomy were treated on SmPF (Figure lA,B); the target volume consisted of the proximal half of the vagina, and the parametrial tissues up to the S1-S2 junction. Patients with positive pelvic lymph

Figure 1. Example of EBRT fields; SmPF (A,B) and StPF (C,D)

115

Chapter 7

116

nodes or inadequate lymphadenectomy were irradiated on StPF (Figure lC,D); the target volume consisted of the proximal half of the vagina, parametrial tissues, internal and proximal external iliac lymph node region and distal part of the common iliac lymph node region. EF was used in case of proven common iliac or peri-aortic lymph node metastases; the target volume consisted of the StPF with 2 cm extension above the positive lymph node (cranial border: up to the Th12-Thll interface). Radiotherapy was given 4-5 times a week using a four-field box standard technique or 3D CT planning. Patients were treated in supine position with full bladder. The dose fraction was 1 .5, 1 .8 or 2.0 Gy and the total dose given ranged from 45 to 50.4 Gy. Dose fractions of 1 .5 Gy were used in 6 patients treated with EF, to reduce the risk of toxicity. External boost radiotherapy was given to the upper vagina or parametrium in case of incomplete or marginal tumor removal during surgery (n=6) or to positive lymph nodes with extra nodal growth (n=2). The total doses of external radiotherapy boost ranged from 6 to 16.2 Gy using dose fractions of 1 .8 Gy or 2.0 Gy. Brachytherapy boost was given in case of cervical involvement (n=lO) or incomplete tumor removal in the vaginal cuff (n=l) and was delivered with a vaginal cylinder, covering the proximal half of the vagina. The dose was specified at 5 mm distance from the surface of the cylinder. Dose schedules were predescribed with low-dose-rate or pulse-dose-rate in 4 patients; total doses were given in one fraction and ranged from 9-20 Gy. Dose schedules were predescribed with high-dose-rate in 7 patients; total doses (6-8 Gy) were given in one fraction (n=6) or in 2 fractions of 5.4 Gy (n=l).

Follow-up Patients were seen for follow-up every 3 months in the first year and every 6 months until 5 years alternating by the gynecologist and radiation oncologist. At each followup visit a history was obtained with special attention for treatment-related morbidities such as toxicity of gastrointestinal tract, urinary tract, lymphedema of limbs or genitals and sexual side effects (vaginal dryness and dyspareunia). Next, a physical and pelvic examination was performed.

Toxicity Toxicity was retrospectively scored by the Common Terminology Criteria for Adverse Events (CTCAE) v3.0 (http:/ /ctep.cancer.gov). Adverse events were scored as acute when they occurred during radiotherapy and as late when they occurred or worsened 3 months after radiotherapy. The grade of adverse event depicted was the highest observed during follow-up.

Data collection and institutional review board approval Data regarding patient, tumor and treatment characteristics were retrieved from an existing database on EC patients and transferred into a separate anonymous database. Inquiries on late toxicity and disease status of patients were part of routine follow-up care. Because of these procedures, according to Dutch law, no further patient or institutional review board approval was needed. Toxicity was determined by reviewing the clinical charts. Data were collected up to July 2010.



percentile) were presented. Chi-square tests (or Fisher 's exact tests, if appropriate) or Mann-Whitney U tests were used to compare clinicopathological characteristics, treatment details and adverse events between treatment groups. The incidence of late events was estimated cumulatively; Kaplan-Meier curves were made and log-rank tests were used for time-to-event analysis with time to complication taken from last day of radiotherapy to the first day of late event. Censoring was done at the date of last contact or death in case of no complication. Complications occurring after recurrent disease were not included, but censored 3 months prior to date of diagnosis of recurrent disease. Associations between clinicopathological parameters and adverse events were estimated using univariate logistic regression analyses. Clinicopathological characteristics and treatment details (being age, BMI, extend of lymphadenectomy, daily dose and total dose) were used as independent and adverse events were used as dependent variables. Odds Ratios (ORs) and 95%-confidence intervals (95%-Cls) were estimated using logistic regression analyses. Tests were performed two-sided and p-values <0.05 were considered statistically significant. Analyses were performed using the software package SPSS, version 16.0 for Windows (SPSS Inc. Chicago Illinois, USA).

Results

Patients In total, 118 EC patients were preoperatively diagnosed with high-grade histology. EBRT was indicated in 84 patients; 75 were evaluable for late toxicity and 72 for acute toxicity. Clinicopathological characteristics, treatment details and outcomes are depicted in Tables 1 and 2, respectively. Median follow-up after radiotherapy during which late toxicity could be evaluated was 33 months (IQR: 18-55). Clinicopathological characteristics (age, BMI, tumor grade, tumor type, myometrial invasion and lymphvascular space involvement) were not significantly different between treatment groups. StPF EBRT was applied in 15 patients (60.0%) despite negative lymph nodes due to inadequate pelvic lymphadenectomy (9 patients) or preference of the treating institute (6 patients). Recurrent disease developed in 18 of 75 patients (24.0%); which were predominantly distant metastases (88.8%). Median time until recurrence was 18.0 months (IQR: 12.8-30.8). Death due to EC occurred more often among patients treated with EF (35.7%) compared to SmPF (6.1 %) and StPF EBRT (17.9%) (p<0.001 and p=0.018, respectively).

117

Table 1. Patient characteristics and pathological features of 75 EC patients treated with EBRT.

All patients (n=75) SmPF (n=33) p-value 1 StPF (n=28) p-value 2 EF (n=15) p-value 3

Age 0.597 0.452 0.230 Median (IQR) 64 (57-73) 67 (61-73) 64 (54-73) 62 (49-70)

BMI 0.691 0.592 0.262 Median (IQR) 27.7 (23.9-30.8) 28.9 (23.7-31 .4) 26.7 (24.1 -31.5) 27.4 (21.9-29.0)

Stage <0.001 0.001 <0.001

I 26 (34.7%) 17 (51.5%) 9 (32.1 %) 0 (0%) II 21 (28.0%) 16 (48.5%) 5 (17.9%) 0 (0%) III 28 (37.3%) 0 (0%) 14 (50.0%) 14 (100%)

Tumor grade 0.336 0.729 0 .263 Grade 1 (well differentiated 6 (8.0%) 0 (0%) 5 (17.9%) 1 (7.1 %) Grade 2 (moderate differentiated) 13 (17.3%) 6 (18.2%) 3 (10.7%) 4 (28.6%) Grade 3 (poorly differentiated) 31 (41.3%) 14 (42.4%) 13 (46.4%) 4 (28.6%) Undifferentiated 2 (3.6%) 1 (3.0%) 1 (3.6%) 0 (0%) Clear cell, papillary serous, mixed high-grade tumor type 23 (30.7%) 12 (36.4%) 6 (21 .4%) 5 (35.7%)

Tumor type 0.153 0.491 0.668 Endometrioid 48 (64.0%) 18 (54.5%) 21 (75 .0%) 9 (64.3%) Clear cell 12 (16.0%) 6 (18.2%) 2 (7.1 %) 4 (28.6%) Papillary serous 7 (9.3%) 5 (15.2%) 1 (3.6%) 1 (7.1%) Mixed tumors 4 (5.3%) 1 (3.0%) 3 (10.7%) 0 (0%) Undifferentiate 2 (2.7%) 1 (3.0%) 1 (3.6%) 0 (0%) Other (squamous cell / carcinosarcoma) 2 (2.7%) 2 (6.1 %) 0 (0% 0 (0%)

Myometrial invasion 0.660 0.923 0 .724 < 50% 34 (45.3%) 16 (48.5%) 12 (42.9%) 6 (42.9%) � 50 41 (54.7%) 17 (51.5%) 16 (57.1 %) 8 (57.1 %)

LVSI 0 .065 0.735 0.084 Negative 47 (64.4% 25 (75.8%) 15 (55.6%) 7 (50.0%) Positive 26 (35.6%) 7 (21.2%) 12 (44.4%) 7 (50.0%) Missing 2 1 1 0

Recurrent disease 0.335 0.298 0.061 No 57 (76.0%) 28 (84.8%) 21 (75.0%) 8 (57.1 %) Yes 18 (24.0%) 5 (15.2%) 7 (25.0%) 6 (42.9%) Loco regional 2 (11.1 %) 1 (20.0%) 0 (0%) 1 (16.7%) Distant 16 (88.9%) 4 (80.0%) 7 (100%) 5 (83.3%)

Follow-up 0.061 0.018 <0.001

Alive 55 (73.3%) 28 (84.8%) 19 (67.9%) 8 (57.2%) Death due to disease 12 (16.0%) 2 (6.1 %) 5 (17.9%) 5 (35.7%) Death due to other disease 8 (10.7%) 3 (9.1 %) 4 (14.3%) 1 (7.1 %)

1 Comparison between SmPF and StPF. 2 Comparison between StPF and EF. 3 Comparison between SmPF and EF. 4 Missing numbers including patients with clear cell, papillary serous tumor type. SmPF: small pelvic field; StPF: Standard pelvic field; EF: extended field ; LVSI: lymphvascular space involvement. Bold values signiftJ p< 0.05.

Table 2. Treatment details of 75 high-risk EC patients.

All patients (n=75) SmPF (n=33) p-value 1 StPF (n=28) p-value 2 EF (n=14) p-value 3

Type of sttrgertJ 0.004 0 .109 0 .710 TAH + BSO 3 (4 .0%) 0 (0%) 3 (10.7%) 0 (0%) TAH + BSO + pelvic lymph nodes 25 (33.3%) 7 (21.2%) 14 (50.0%) 4 (28.6%) TAH + BSO + pelvic- and para-aortic 47 (62.7%) 26 (78 .8%) 11 (39 .3%) 10 (71.4%)

N of pelvic lymph nodes harvested 0.027 0 .515 0.008 1-9 13 (18.0%) 0 (0%) 9 (36.0%) 4 (28 .6%) 10-19 39 (54 .2%) 21 (63 .6%) 9 (36 .0%) 9 (64 .3%) ;;;:20 20 (27.8%) 12 (36.4%) 7 (28 .0%) 1 (7 .1 %) Median number of lymph nodes (IQR) 15 (10-20) 17 (13-23) 13 (8-20) 12 (9-17)

Involvement of pelvic nodes <0.001 0.001 <0.001 No 49 (68.1%) 33 (100%) 15 (60.0%) 1 (7 .1%)4) Yes 23 (31.9%) 0 (0%) 10 (40 .0%) 13 (92.9%)

N of para-aortic lymph nodes harvested 0.477 0.557 0.986 1-2 19 (40.4%) 11 (42 .3%) 2 (20.0%) 5 (50 .0%) ;;;:3 28 (59.6%) 15 (57. 7%) 8 (80.0%) 5 (50.0%) Median number of lymph nodes (IQR) 3 (2-4) 3 (1-5) 3 (2-3) 3 (2-3)

Involvement of para-aortic nodes 0.297 <0.001 <0.001 No 37 (78 .7%) 26 (100%) 10 (90.9%) 1 (10.0%) Yes 10 (21 .3%) 0 (0%) 1 (9.1%)5) 9 (90 .0%)

Radiotherapy: fraction dose 1.5 Gy 6 (8 .0%) 0 (0%) 0.037 0 (0%) 0.001 6 (42 .9%) <0.001 1.8 Gy 26 (34. 7%) 7 (21.2%) 13 (46 .4%) 6 (42.9%) 2.0 Gy 43 (57 .3%) 26 (78.8%) 15 (53 .6%) 2 (14.2%)

Boost 0 .107 0.172 0.009 No 56 (74.7%) 29 (87 .9%) 20 (71.4%) 7 (50.0%) Yes 19 (25 .3%) 4 (12 .1 %) 8 (28 .6%) 7 (50.0%)

Type of boost radiotherapy 1 .000 0.132 0.005 Brachytherapy 11 (57.9%) 3 (75.0%) 6 (75 .0%) 2 (28.6%) External radiotherapy 8 (42.1 %) 1 (25 .0%) 2 (25.0%) 5 (71.4%)

1 Comparison between SmPF and StPF. 2 Comparison between StPF and EF. 3 Comparison between SmPF and EF. 4 patient had isolated para-aortic lymph node metastases. 5 patient had metastasis in lymph node located low para-aortic and was treated with EBRT on StPF with caudal border L5-L4. Bold values signifiJ p< 0.05. SmPF: small pelvic field; StPF: Standard pelvic field; EF: extended field.

Table 3. Acute adverse events reported in patients treated with EBRT

All cases (n=75) SmPF (n=33) p-value 1 StPF (n=28) p-value 2 EF (n=14) p-value 3

Data missing 3 1 1 1 Total patients with toxicittJ: 52 (72.2%) 21 (65.6%) n.s. 19 (70.3%) n.s. 12 (92.3%) n.s.

Gastrointestinal 41 (56.9%) 14 (43.8%) n.s. 16 (59.3%) n.s. 11 (84.6%) 0.012 Anorexia/Nausea 16 (22. 2%) 2 (6.1 %) n.s. 5 (16.7%) 0.004 11 (68.8%) <0.001 Diarrhea 37 (51.4%) 12 (36.4%) n.s. 15 (20.8%) n.s. 12 (75.0%) 0.016 Pain/irritation anal 5 (6.9%) 1 (3.0%) n.s. 3 (10.0%) n.s. 1 (6.3%) n.s. Pain, abdominal 2 (2.8%) 2 (6.1 %) n.a. 0 (0%) n.a. 0 (0%) n.a.

Urinary tract 12 (16.7%) 8 (25.0%) n.s. 4 (14.8%) n.s. 0 (0%) n.s. Cystitis 6 (8.3%) 6 (18. 2%) n.s. 1 (3.3%) n.s. 1 (6.3%) n.s. Frequency/urgency 5 (6.9%) 2 (6.1 %) n.s. 6 (20.0%) n.s. 0 (0%) n.s. I ncontinence, urinary 1 (1.4%) 1 (3.0%) n.s. 0 (0%) n.a. 0 (0%) n.s.

Fatigue / 3 (4. 2%) 1 (3.0%) n.s. 2 (6.7%) n.s. 0 (0%) n.s. Deep venous thrombosis 1 (1 .4%) 0 (0%) n.a. 1 (3.3%) n.a. 0 (0%) n.a.

1 Comparison between SmPF and StPF. 2 Comparison between StPF and EF. 3 Comparison between SmPF and EF. 4 Functional obstruction of the bowel SmPF: small pelvic field; StPF: Standard pelvic field; EF: extended field ; n.s. not significant; n.a. not applicable

Table 4. Late adverse events reported in patients treated with EBRT.


All sites 53 (70.7%) 20 (60.6%) n.s. 20 (71.4%) n.s. 13 (92.9%) n.s.

Gastrointestinal 36 (48.0%) 11 (33.3%) 0.032 17 (60.7%) n.s. 8 (57.1%) n.s.

Anorexia/Nausea 11 (14.7%) 1 (3.0%) 0.004 9 (32.1 %) n.s. 1 (7.1 %) n.s.

Diarrhea 20 (26.7%) 6 (18.2%) n.s. 8 (28.6%) n.s. 6 (42.9%) n.s.

Dyspepsia 1 (1.3%) 0 (0%) n.a. 0 (0%) n.s. 1 (7.1 %) n.s.

Enteritis 11 (14.7%) 3 (9.1 %) n.s. 6 (21 .4%) n.s. 2 (14.3%) n.s.

Flatulence 2 (2.7%) 2 (6.1 %) n.s. 0 (0%) n.a. 0 (0%) n.s.

Hemorrhage 6 (8.0%) 1 (3.0%) n.s. 3 (10.7%) n.s. 2 (14.3%) n.s.

Ileus 4 5 (6.7%) 0 (0%) 0.039 4 (14.3%) n.s. 1 (7.1 %) n.s.

Incontinence, anal 11 (14.7%) 5 (15.2%) n.s. 5 (17.9%) n.s. 1 (7.1 %) n.s.

Pain, abdominal 2 (2.7%) 1 (3.0%) n.s. 0 (0%) n.s. 1 (7.1 %) n.s.

Proctitis 3 (4.0%) 1 (3.0%) n.s. 1 (3.6%) n.s. 1 (7.1 %) n.s.

Stricture/ s tenosis 3 (4.0%) 1 (3.0%) n.s. 1 (3.6%) n.s. 1 (7.1 %) n.s.

UrinanJ tract 22 (29.3%) 9 (27.3%) n.s. 9 (32.1%) n.s. 4 (28.6%) n.s.

Hemorrhage 1 (1 .3%) 0 (0%) n.a. 0 (0%) n.s. 1 (7.1 %) n.s .

Cystitis 6 (8.0%) 3 (9.1 %) n.s. 1 (3.6%) n.s. 2 (14.3%) n.s.

Frequency/urgency 10 (13.3%) 6 (18.2%) n.s. 2 (7.1 %) n.s. 2 (14.3%) n.s.

Incontinence, urinary 14 (18.7%) 4 (12.1 %) n.s. 7 (25 .0%) n.s. 3 (21 .4%) n.s.

Stricture/stenosis 2 (2.7%) 2 (6.1 %) n.s. 0 (0%) n.a. 0 (0%) n.s.

Lymphedema 18 (24.0%) 6 (18.2%) n.s. 8 (28.6%) n.s. 4 (28.6%) n.s.

Limbs 16 (21.3 %) 6 (18.2%) n.s. 7 (25 .0%) n.s. 3 (21 .4%) n.s.

Genital 4 (5 .3%) 1 (3.0%) n.s. 2 (7.1 %) n.s. 1 (7.1 %) n.s.

Fatigue 15 (20.0%) 6 (18.2%) n.a. 6 (21 .4%) n.s. 3 (21.4%) n.s.

Deep venous thrombosis 1 (1.3%) 1 (3.0%) n.a. 0 (0%) n.a. 0 (0%) n.s.

Vaginal bleeding 1 (1.3%) 0 (0%) n.a. 0 (0%) n.s. 1 (7.1 %) n.s.

1 Comparison between SmPF and StPF. 2 Comparison between StPF and EF. 3 Comparison between SmPF and EF. 4 Functional obstruction of the bowel.

SmPF: small pelvic field; StPF: Standard pelvic field; EF: extended field; n.s. not significant; n.a. not applicable Bold values signifiJ p< 0.05.

Table 5. Highest grade of late toxicity reported.


All sites 53 (70.7%) 20 (60.6%) 20 (71.4%) 13 (92.9%)

Grade 1 34 (45.4%) 13 (39.4%) n.s. 13 (46.4%) n.s. 8 (57.2%) n.s. Grade 2 13 (17.3%) 4 (12.1 %) 5 (17.8%) 4 (28.6%) Grade 3 5 (6.7%) 3 (9. 1 %) 1 (3.6%) 1 (7.1 %) Grade 4 1 (1 .3%) 0 (0%) 1 (3 .6%) 0 (0%)

Gastrointestinal 37 (49.3%) 12 (36.4%) 17 (60.7%) 8 (57.1%)

Grade 1 24 (32.0%) 8 (24.3%) n.s. 10 (35 .7%) n.s. 6 (42.9%) n.s. Grade 2 9 (12.0%) 3 (9.1 %) 5 (17.8%) 1 (7.1 %) Grade 3 3 (4.0%) 1 (3.0%) 1 (3.6%) 1 (7.1 %) Grade 4 1 (1.3%) 0 (0%) 1 (3.6%) 0 (0%0

UrinanJ tract 22 (29.3%) 9 (27.3%) 9 (32.1%) 4 (28.6%)

Grade 1 13 (17.3%) 4 (12.1 %) n.s. 8 (28.5%) n.s. 1 (7.1 %) n.s. Grade 2 7 (9.3%) 3 (9.1 %) 1 (3.6%) 3 (21.4%) Grade 3 2 (2.7%) 2 (6.1 %) 0 (0%) 0 (0%) Grade 4 0 (0%) 0 (0%) 0 (0%) 0 (0%)

Lymphatics 18 (24.0%) 6 (18.2%) 8 (28.6%) 4 (28.6%)

Grade 1 17 (22.7%) 5 (15.2%) n.s. 8 (28.6%) n.s. 4 (28.6%) n.s. Grade 2 1 (1 .3%) 1 (3.0%) 0 (0%) 0 (0%) Grade 3 0 (0%) 0 (0%) 0 (0%) 0 (0%) Grade 4 0 (0%) 0 (0%) 0 (0%) 0 (0%)

Fatigue 15 (20.0%) 6 (18.2%) 6 (21.4%) 3 (21.4%)

Grade 1 14 (18.7%) 5 (15.2%) n.s. 6 (21.4%) n.s. 3 (21.4%) n.s. Grade 2 1 (1.3%) 1 (3.0%) 0 (0%) 0 (0%) Grade 3 0 (0%) 0 (0%) 0 (0%) 0 (0%) Grade 4 0 (0%) 0 (0%) 0 (0%) 0 (0%)

1 Comparison between SmPF and StPF. 2 Comparison between StPF and EF. 3 Comparison between SmPF and EF. 4 Functional obstruction of the bowel. SmPF: small pelvic field; StPF: Standard pelvic field; EF: extended field; n.s. not significant


Toxicity Acute adverse events were reported in 52 patients (72.2%) (Table 3). Diarrhea was the most common complication (51.4%) and occurred more frequently among patients treated with EF (75.0%) compared to SmPF (36.4%) (p=0.016). Nausea or anorexia occurred in 22.2% and was more frequently reported among patients treated with EF (68.8%) compared to SmPF (6.1 %) and StPF (16.7%) (p=0.004 and p<0.001, respectively). No grade 3 toxicity was reported. Late adverse events were reported in 53 of 75 patients (70.7%) (Table 4). The majority of these patients also reported toxicity during radiotherapy (71 %). Late adverse events of the gastrointestinal tract were the most common (48.0%). The probability of developing a gastrointestinal adverse event was higher among patients treated with StPF (n=l7, 60.7%) compared to SmPF (n=ll, 33.3%) (p=0.040) (Figure 2). Nausea and anorexia more often occurred among patients treated with StPF compared to SmPF (p=0.004). lieus (small bowel obstruction) was not reported in the SmPF group in contrast to StPF treated patients (n=4, 14.3%); two of these patients needed surgery to resolve the ileus due to stenosed bowel segments or extensive adhesions. One patient in the extended group received similar surgery for ileus. All these patients had a medical history of previous abdominal surgery in contrast to the patients in whom ileus could be resolved with conservative treatment. Table 5 summarizes toxicity according to the highest grade of late adverse events reported during follow-up. One patient reported a grade 4 adverse event after treatment with StPF and experienced severe, uncontrollable diarrhea and incontinence for faeces requiring permanent bowel diversion. Overall, no significant difference was found in grade of toxicity between treatment groups.

::Y�EF::: +- - - - - - - - - - - -+

' + - - - - - - - ..... ... - - - ... - - + .. - - -

- - - -·

w � � � � � follow-up time after radiotherapy (months)

Figure 2. Cumulative probabilihJ of late gastrointestinal adverse events, comparison between Small pelvic field (SmPF) and Standard pelvic field (StPF) (p=0.040, Log-rank)

123

Chapter 7

124

Clinicopathological characteristics and adverse events In univariate logistic regression analysis, age and BMI were associated with increased risk for developing ileus (OR: 0.90; 95%-CI: 0.81-0.99, p=0.026 and OR: 0.78; 95%-CI: 0.62-0.98, p=0.036, respectively). In multivariate logistic regression analysis with both variables included in the model, no independent association was found between age, BMI and ileus (age: OR: 0.91; 95%-CI: 0.82-1.01, p=0.085 vs. BMI: OR: 0.82; 95%-CI: 0.66-1.03, p=0.091).

Discussion

In this retrospective study we evaluated toxicity in EC patients treated with TAH+BSO, lymphadenectomy and adjuvant EBRT, and especially focused on late toxicity. We showed that EBRT on SmPF (including the central pelvis and proximal vagina) results in less late gastrointestinal symptoms, especially nausea, anorexia and ileus, compared to treatment with StPF.

The most common late adverse events after adjuvant EBRT were related to the gastrointestinal tract, in agreement with previous studies assessing radiotherapy-related toxicity in EC patients.3•5•7 The (small) bowel is rather sensitive to irradiation and therefore the most vulnerable organ in the pelvis. Bowel toxicity includes diarrhea, enteritis (abdominal pain and blood/mucus in the faeces) and ileus. In order to reduce radiotherapy-related toxicity without compromising central pelvic disease control, we treated patients without lymph node metastases with SmPF EBRT. More specifically, the target volume of SmPF only contained the central pelvis and upper vagina. Consequently, small bowel volume was reduced compared to StPF. Late gastrointestinal adverse events presented significantly more often in patients treated with StPF EBRT compared to patients treated with SmPF EBRT, especially nausea, anorexia and ileus. In accordance with our data, a previous study in cervical cancer patients comparing the use of postoperative radiotherapy on a SmPF to "the classic whole pelvic field" showed that ileus occurred more frequently in the whole pelvic group (16.2%) compared to the small pelvic group (3.2%).21 This study did not analyze nausea and anorexia separately as late adverse event. As expected, the adverse events of the urinary tract were similar between treatment groups. Patients treated with TAH+BSO and lymphadenectomy have a higher risk for early and postoperative complications such as ileus and lymphedema compared to treatment with TAH+BSO only.10•11 Furthermore, morbidity rates further increase with the extend of lymphadenectomy and the additional use of radiotherapy.6•22•23

We observed that lymphedema occurred in a significant proportion of the patients (24.0%) which is in agreement with percentages found in two studies among EC patients treated with TAH+BSO, lymphadenectomy and radiotherapy (37.8% and 27.8%).22•23

Furthermore, lymphedema did not differ between treatment groups, in agreement with the study in cervical cancer where lymphedema occurred in 38.5% of the SmPF and 37.5 % of the StPF group.21 The combination of TAH+BSO, lymphadenectomy and


EBRT might also explain the relatively high frequency of ileus reported in our population (14.3% in StPF and 7.1% in the EF group). Compared to literature, the percentage of patients reporting nausea or anorexia was quite high.7 However, according to the CTCAE v3.0, all patients reported mild (grade 1) to moderate complications (grade 2).

In our study population, patients in the SmPF group underwent more extensive surgery than the StPF group; lymphadenectomy was more frequently performed and the total number of lymph nodes removed was higher. However, we did not find an association between the number of harvested lymph nodes and the occurrence of late adverse events. We acknowledge that extensive surgical staging influences frequency and severity of late toxicity and frequencies reported in our study are therefore higher than the PORTEC 1 results.7 In our population, this is probably equally distributed between treatment groups (or might be even higher in the SmPF group). In 6 patients of the StPF group (21 %), EBRT was applied despite negative lymph nodes after adequate lymphadenectomy based on the preference of the treating institute. Late toxicity was reported in 3/6 patients; gastrointestinal toxicity in 2 patients (grade 1 and 2) and urinary tract toxicity in 1 patient. Despite the latter, above-mentioned facts can not explain that late gastrointestinal toxicity is less frequently observed in the SmPF group compared to the StPF group and is therefore probably based on a volume-based decrease of toxicity.

In univariate regression analysis, younger age and lower BMI were associated with a higher risk of developing ileus after EBRT, but not independently as assessed in multivariate analysis. Where age and BMI were similar between treatment groups, we conclude that these factors did not contribute to the difference of adverse events between SmPF and StPF.

Where gastrointestinal toxicity can occur up to many years after radiotherapy, most patients will develop symptoms during the first 2 years after treatment. In contrast, the development of late toxicity of the urinary tract can occur up to 20 years after radiotherapy.24 Where our median follow-up time was 33 months (IQR: 18-55) the majority of gastrointestinal adverse events are probably detected.

A limitation of our study is the relatively small number of patients that could be analyzed, next to the fact that scoring and grading of adverse events was done by retrospective evaluation of patient charts. This may have caused a certain degree of interobserver variation and grading of mild complications. Severe adverse events, especially those requiring intervention, are always reported in patient charts and are therefore a true reflection of existing moderate/severe (grade 2, 3 or 4) adverse events. It is possible that mild symptoms are not always reported by the patients or noted on the chart by the physician.

Based on the results of our study, it would be interesting to determine whether patients with high-risk EC could be safely treated with VBT or other techniques (intensitymodulated radiation therapy) to further reduce radiotherapy-associated toxicity and without compromising locoregional recurrences. The PORTEC-2 trial has shown that in patients with high-intermediate risk EC, VBT can be used to optimize local control

125

Chapter 7

126

with less severe side effects and better quality of life.5•9 To our knowledge, only retrospective studies have addressed this issue in high-risk EC patients 25•26 and prospective randomized trials are therefore warranted. In this respect, the optimal target volume needs to be determined. In our population, recurrent disease occurred slightly more frequent among the EF (42.9%) and StPF (25.0%) compared to the SmPF group (15.2%), although differences were not significant. In general, we assume that locoregional control was quite good where only 2 patients had recurrent disease in the pelvic area (1 in the SmPF group and 1 in the EF group). However, distant metastases accounted for recurrent disease in the majority of the cases (88.8%) which is a well-known problem among high-risk EC patients. Therefore, more research is needed in order to optimize treatment strategies, possibly by the addition of chemotherapy to radiotherapy. Hopefully, the outcome of the several trials investigating the role of chemotherapy will help determining its role in high-risk EC patients. In conclusion, EC patients without lymph node metastases after adequate surgical staging, in whom adjuvant EBRT is indicated, can be treated with EBRT limited to SmPF, to reduce the extent of adverse events. Where prognosis of high-risk EC patients is still poor, more prospective studies are warranted in order to optimize treatment.

Acknowledgements

The authors would like to thank Dr. C.L. Creutzberg (Department of Clinical Oncology of the Leiden University Medical Center, Leiden, the Netherlands), for critical review and useful suggestions for this manuscript.

Less gastrointestinal toxicity after adjuvant radiotherapy on a small pelvic field compared to a standard pelvic field in patients with endometrial carcinoma ------------------------ --------------- ------

References

1 . Creutzberg CL, van Putten WL, Koper PC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomized trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355:1404-11.

2. Aalders J, Abeler V, Kolstad P, et al. Postoperative external irradiation and prognostic parameters in stage I endometrial carcinoma: clinical and histopathologic study of 540 patients. Obstet Gynecol 1980;56:419-27.

3. Blake P, Swart AM, Orton J, et al. Adjuvant external beam radiotherapy in the treatment of endometrial cancer (MRC ASTEC and NCIC CTG EN.5 randomised trials): pooled trial results, systematic review, and meta-analysis. Lancet 2009;373:137-46.

4. Keys HM, Roberts JA, Brunetto VL, et al. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 2004;92:744-51.

5. Nout RA, Putter H, Jurgenliemk-Schulz IM, et al. Quality of life after pelvic radiotherapy or vaginal brachytherapy for endometrial cancer: first results of the randomized PORTEC-2 trial. J Clin Oncol 2009;27:3547-56.

6. Corn BW, Lanciano RM, Greven KM, et al. Impact of improved irradiation technique, age, and lymph node sampling on the severe complication rate of surgically staged endometrial cancer patients: a multivariate analysis. J Clin Oncol 1994;12:510-5.

7. Creutzberg CL, van Putten WL, Koper PC, et al. The morbidity of treatment for patients with Stage I endometrial cancer: results from a randomized trial. Int J Radiat Oncol Biol Phys 2001;51:1246-55.

8. Huscher A, Bignardi M, Magri E, et al. Determinants of small bowel toxicity in postoperative pelvic irradiation for gynaecological malignancies. Anticancer Res 2009;29:4821-6.

9. Nout RA, Smit VT, Putter H, et al. Vaginal brachytherapy versus pelvic external beam radiotherapy for patients with endometrial cancer of high-intermediate risk (PORTEC-2): an openlabel, non-inferiority, randomised trial. Lancet 2010;375:816-23.

10. Benedetti Panici P, Basile S, Maneschi F, et al. Systematic pelvic lymphadenectomy vs. no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst 2008;100:1707-16.

11. Kitchener H, Swart AM, Qian Q, et al. Efficacy of systematic pelvic lymphadenectomy in endometrial cancer (MRC ASTEC trial): a randomised study. Lancet 2009;373:125-36.

12. Hidaka T, Kato K, Yonezawa R, et al. Omission of lymphadenectomy is possible for low-risk corpus cancer. Eur J Surg Oncol 2007;33:86-90.

13. Mariani A, Webb MJ, Keeney GL, et al . Low-risk corpus cancer: is lymphadenectomy or radiotherapy necessary? Am J Obstet Gynecol 2000;182:1506-19.

14. Cragun JM, Havrilesky LJ, Calingaert B, et al. Retrospective analysis of selective lymphadenectomy in apparent early-stage endometrial cancer. J Clin Oncol 2005;23:3668-75

15. Chan JK, Cheung MK, Huh WK, et al. Therapeutic role of lymph node resection in endometrioid corpus cancer: a study of 12,333 patients. Cancer 2006;107:1823-30.

16. Lutman CV, Havrilesky LJ, Cragun JM, et al . Pelvic lymph node count is an important prognostic variable for FIGO stage I and II endometrial carcinoma with high-risk histology. Gynecol Oncol 2006;102:92-7.

127

17. Todo Y, Kato H, Kaneuchi M, et al. Survival effect of para-aortic lymphadenectomy in endometrial cancer (SEPAL study): a retrospective cohort analysis. Lancet 2010;375:1165-72.

18. Bernardini MQ, Murphy JK. Issues surrounding lymphadenectomy in the management of endometrial cancer. J Surg Oncol 2009;99:232-41.

19. Kridelka FJ, Berg DO, Neuman M, et al. Adjuvant small field pelvic radiation for patients with high risk, stage 1B lymph node negative cervix carcinoma after radical hysterectomy and pelv ic lymph node dissection. A pilot study. Cancer 1999;86:2059-65.

20. Nijman HW, Khalifa M, Covens.A. What is the number of lymph nodes required for an "adequate" pelvic lymphadenectomy? Eur J Gynaecol Oncol 2004;25:87-9.

21. Ohara K, Tsunoda H, Satoh T, et al. Use of the small pelvic field instead of the classic whole pelvic field in postoperative radiotherapy for cervical cancer: reduction of adverse events. I nt J Radiat Oneal Biol Phys 2004;60:258-64.

22. Tada H, Teramukai S, Fukushima M, et al. Risk factors for lower limb lymphedema after lymph node dissection in patients with ovarian and uterine carcinoma. BMC Cancer 2009;9:47.

23. Todo Y, Yamamoto R, Minobe S, et al. Risk factors for postoperative lower-extremity lymphedema in endometrial cancer survivors who had treatment including lymphadenectomy. Gynecol Oncol 2010;119:60-4.

24. Maduro JH, Pras E, Willemse PH, et al. Acute and long-term toxicity following radiotherapy alone or in combination with chemotherapy for locally advanced cervical cancer. Cancer Treat Rev 2003;29:471-88.

25. McCloskey SA, Tchabo NE, Malhotra HK, et al. Adjuvant vaginal brachytherapy alone for high risk localized endometrial cancer as defined by the three major randomized trials of adjuvant pelvic radiation. Gynecol Oncol 2010;116:404-7.

26. Siow T R, Yeo MC, Khoo-Tan HS, et al. Stage lC Grade 3 Endometrial Cancer: T he KK Hospital Gynaecological Oncology Group Experience. I nt J Gynecol Cancer 2010;20:1557-62.

Chapter 8

Summarizing discussion and

future perspecti'-qes

Summarizing discussion and future perspectives

Summarizing discussion

Endometrial carcinoma (EC) is a heterogeneous disease with several histological types and different biological behavior. The general prognosis for EC is good, but in patients diagnosed with non-endometrioid subtypes such as serous carcinoma, clearcell carcinoma or carcinosarcomas, life-expectancy decreases dramatically. Currently, patients with low-risk EC ( <50% myometrial invasion, endometrioid tumor type, tumor grade 1-2) are solely treated with surgery, consisting of hysterectomy and bilateral salpingooophorectomy (BSO). In case of EC with high-intermediate risk factors, surgery is followed by radiotherapy.1 Over the last years, efforts have been undertaken to explore new treatment modalities and optimize standard treatments for EC in an effort to reduce side-effects on one hand and increase efficacy on the other. Immunotherapy is an interesting new treatment option where it is generally recognized that the immune system has an important role in regulating cancer development and progression.2

This thesis mainly focuses on aspects of the immune system in EC. Next, molecular markers and clinical behavior of uterine carcinosarcomas (UCS) were studied in order to improve knowledge on this rare subtype of EC. In addition, toxicity after treatment of patients with high-risk EC was evaluated.

Role of the immune system in EC The interaction between the immune system and cancer cells has been subject of investigation for several years. The observations of tumor-specific immunity and the prognostic influence of tumor-infiltrating lymphocytes (TIL)3•6 paved the way for developing immunotherapy as new treatment modality for cancer. Compared to other cancer types, knowledge on the role of the immune system in EC has lagged behind. To this end, chapter 2 evaluated the prognostic influence of three TIL subtypes in primary tumor tissue of 368 EC patients. CD8+ cytotoxic T-lymphocytes, Foxp3+ regulatory Tlymphocytes and CD45RQ+ memory T-lymphocytes were present in 96.8%, 79.4% and 60.7% of the tumor specimens, respectively. The presence of high numbers of CD8+

T-lymphocytes was predictive for improved survival in the entire cohort of EC patients and in a subgroup of patients with type II EC (non-endometrioid histology). Furthermore, overall survival in type II EC was positively influenced by the presence of CD45R0+ memory T-lymphocytes. Foxp3+ regulatory T-lymphocytes did not influence survival. The importance of a balance between effector and suppressor immune cells was outlined by the fact that a high CD8+ /Foxp3+ T-lymphocyte ratio was associated with favorable prognostic factors and longer disease-free survival (DFS). The favorable influence of CD8+ cytotoxic T-lymphocytes and CD45Ro+ memory T-lymphocytes on survival indicates that tumor-specific immunity is present in EC and that immunotherapy might be introduced as new treatment modality in EC.

Immunotherapy aims to potentiate tumor-specific immunity in order to eradicate cancer cells or prevent further tumor outgrowth. Given the fact that tumor mass is still present despite the presence of CTLs in the tumor microenvironment, the immune system is not able to eradicate all cancer cells in these cases. In fact, cancer cells are able to escape from immune-mediated recognition and killing via several mechanisms. In-

131

Chapter 8

132

depth knowledge of escape mechanisms is essential in order to design an (effective) immunotherapy-based treatment modality. One such escape-mechanism is downregulation of MHC class I, associated with worse survival in EC patients.7

Studies in colorectal carcinoma demonstrated that downregulation of MHC class I more frequently occurs in tumors with an aberrant mismatch repair (MMR) system. In general, cancers with a deficient (MMR) system elicit strong and specific cellular immune responses through increased production of immunogenic peptides. However, a subsequent selective pressure favors the outgrowth of tumor cell clones that have lost MHC class I expression enabling cancer cells to impair CTL-mediated recognition and killing of cancer cells.8-10 In order to determine if such a mechanism also applies for EC, expression of 3 important MMR proteins (MLHl, MSH2 and MSH6) and MHC class I was determined in tumor specimens of 486 patients with sporadic (i.e. non-hereditary) endometrioid EC. As chapter 3 demonstrated, loss of MMR protein expression correlated with concurrent MHC class I downregulation. As expected, tumors without MHC class I expression contained lower numbers of intratumoral CDS+ cytotoxic Tlymphocytes compared to tumors with MHC class I expression. Moreover, patients with loss of MMR protein expression had worse survival rates compared to patients with expression of all MMR proteins. A high number of CDS+ T-lymphocytes had a positive influence on survival among patients with expression of all MMR proteins in contrast to ECs without MMR protein expression. These results imply that EC patients with an abnormal MMR system more frequently have downregulation of MHC class I and as a consequence, contain less tumor-specific CTLs. Hence, these characteristics enable tumors to escape from specific immune responses which results in worse survival rates for patients. Therapies aiming at induction of MHC class I expression might be especially beneficial in patients without MMR protein expression and/ or MHC class I expression. Candidates for this strategy could be selected by determining MMR and MHC class I status in tumor specimens acquired from surgery. In addition, therapeutic strategies leading to an additional increase in the number of tumor-specific CTLs might add to clinically effective anti-tumor immune responses in EC.

In addition to MHC class I downregulation, several other mechanisms exist by which cancer cells can circumvent an immune-attack. In this respect, increased production of indoleamine 2,3-dioxygenase (IDO) as cancer-induced escape mechanism is subject of investigation in several cancer types.11 IDO is an intracellular enzyme involved in the breakdown of the essential amino-acid tryptophan which is essential for protein and indole synthesis. Tryptophan is metabolized along the kynurenine pathway where several downstream metabolites are formed such as kynurenine and 3-hydroxykynurenine and nicotinamide adenine dinucleotide (NAD+), ultimately. In serum, IDO activity can be estimated by the kynurenine-to-tryptophan ratio. In chapter 4, IDO activity was determined in serum of 109 patients with endometrial, ovarian and vulvar cancer by an automated on-line solid-phase extraction-liquid chromatographic-tandem mass spectrometric (XLC-MS/MS) method. Cancer patients had enhanced tryptophan degradation which resulted in higher serum concentrations of kynurenine and a higher kyn/ trp ratio as compared to healthy controls. These findings are probably based on higher systemic IDO activity among cancer patients. Studies in several cancer types have demonstrated that IDO expression in tumor tissue


is associated with poor prognosis.12-16 The hazardous effect of IDO is due to its suppressive effect on effector T-lymphocytes; due to depletion of tryptophan (amongst others, see chapter 1), any T-lymphocyte or natural killer (NK) cell in this micro-environment is more prone to cell cycle arrest and apoptosis. !DO-induced suppression of T-lymphocytes can be reversed by 1-methyltryptophan (1-MT) which is a pharmacological inhibitor of IDO. In order to predict if treatment with !DO-blocking agents would be beneficial for EC patients, the prognostic role of IDO was determined in chapter 5. As expected, tumor tissue with IDOhigh expression contained lower numbers of CD8+

T-lymphocytes compared to tumors with IDO10w expression.12•14•1 7 Moreover, IDOhigh expression was independently associated with poor disease specific survival (DSS) in the general cohort of EC patients and among patients with early stage EC. These findings indicate that !DO-blockage might only help to improve DSS but not prevent recurrent disease whereas no effect on DFS was observed.

Before IDO-blocking agents can be introduced as anti-cancer treatment modality, more knowledge is needed on the exact immunosuppressive mechanism of IDO. !DO-induced tryptophan depletion causes cell cycle arrest by triggering the antiproliferative GCN2 pathway in lymphocytes.18 It is unclear why especially T-lymphocytes are sensitive for !DO-induced tryptophan depletion. In fact, cancer cells remain unaffected whilst in the same microenvironment as T-lymphocytes. As described in chapter 5 as well, several experiments using a human cancer cell line confirmed that IDO expression is dependent on IFN-y stimulation.12•14•16 Next, electron microscopy was used to investigate the exact localization and distribution of IDO in cancer cells. As demonstrated for the first time, cytoplasmic IDO expression was predominantly localized at the level of the cell membrane. Measurements by XLC-MS/MS were performed to investigate IDO at a more functional level. Tryptophan depletion predominantly occurred in the extracellular environment of cancer cells while intracellular concentrations changed moderately. The dramatic decrease of extracellular tryptophan is lethal to T-lymphocytes whereas intracellular tryptophan levels are maintained at an acceptable level enabling cancer cells to survive. This might be explained by the sub-cellular localization of IDO, facilitating the maintenance of an adequate intracellular level of tryptophan. Upregulation of the system L amino acid transporter 1 (LATl) in cancer cells is another possible explanation.19•20 LATl is involved in the transport of large amino acids such as tryptophan in the intracellular environment. In addition, a novel tryptophan transporter was recently discovered to be upregulated by IDO in mice and humans.21 Tryptophan transporters might be involved in maintaining an adequate cellular tryptophan status where the cellular environment of the IDO expressing cells become tryptophan depleted. Ideally, blockage of cancer-specific tryptophan transporters as new anti-cancer treatment would lead to restoration of adequate number and function of tumor-specific T-lymphocytes and subsequent killing of cancer cells.

UCS, aggressive subtypes of EC Although new treatments modalities are being explored for EC, more knowledge on clinicopathological characteristics and behavior of EC subtypes is essential before clinical application. This especially holds for aggressive subtypes such as UCS. Patients with UCS are currently treated as high-risk EC but have worse survival rates.22 The

133

Chapter 8

134

fact that UCS only accounts for 1-5% of uterine malignancies, has complicated detailed investigation (e.g. by randomized trials) of this tumor type. Therefore, molecular characteristics and clinical behavior of UCS were investigated in a cohort of 40 patients with UCS, and described in chapter 6. Molecular markers such as p53, MSH2 and MSH6 are important factors in early carcinogenesis. Expression of these markers in the epithelial and mesenchymal malignant components of UCS were determined and confirmed the monoclonal origin of UCS.23•24 Expression patterns of p53, MMR proteins, PTEN and hormone receptors showed high similarities to results previously described in EC.25 These findings indicate that UCS and EC follow similar developmental pathways. In UCS, the epithelial tumor component is the 'driving force' where the majority of metastases (72%) and vascular invasion (70%) were caused by the epithelial tumor component only.26•27

In EC, prognosis and response to treatment depend on the histological subtype (endometrioid or non-endometrioid) and a similar prognostic profile might also apply for UCS. To this end, clinicopathological characteristics and survival were compared between UCS patients with a non-endometrioid epithelial tumor component and an endometrioid epithelial tumor component. Patients with a non-endometrioid component tended to have more peritoneal metastases and worse DSS (5-year survival: 26%) compared to an endometrioid component (5-year survival: 55%). However, DSS was significantly worse for patients with UCS (5-year survival: 42%) compared to EC with high-grade features; non-endometrioid (5-year survival: 57%) and grade 3 endometrioid EC (5-year survival: 77%) underlining the need for improved treatment strategies. The results of chapter 6 suggest that UCS can be divided in tumors with endometrioid and non-endometrioid features, just as EC. As a consequence, treatment strategies for UCS should focus on characteristics of the epithelial tumor component. Due to the rarity of UCS, randomized trials will not be easily performed and it will probably be inevitable to incorporate UCS as high-risk category in future EC trials.

Treatment-related toxicity In addition to the identification of new treatment modalities, it is inevitable to aim for the improvement of standard treatment modalities. Nowadays, surgery is still the cornerstone of EC treatment and radiotherapy is most frequently used as adjuvant treatment modality. EC patients with high-intermediate risk factors have increased risk for recurrent disease compared to patients with low-risk EC (80% of all EC patients). The latter group is treated with surgery only whereas patients with high-intermediate risk factors are treated with surgery followed by adjuvant radiotherapy.1 Radiotherapy is associated with short and long-term morbidity and should therefore merely be used in patients that benefit most. Efforts have been made to reduce treatment-related sideeffects without compromising locoregional disease control of recurrences. Patients with intermediate-high risk EC can be safely treated with vaginal brachytherapy (VBT) instead of external beam radiation therapy (EBRT) while causing significantly less sideeffects and improved quality-of-life.28 Therefore, EBRT is currently indicated only for patients with high-risk EC or advanced disease (FIGO III). In chapter 7, treatment-related toxicity of high-risk EC patients was evaluated. EC patients were treated with surgery including lymphadenectomy based on pre-operatively determined factors (grade 3 endometrioid, non-endometrioid tumor types and/ or cervical invasion). EBRT on a small pelvic field (including only the central pelvis and proximal vagina) was applied


in case of proven negative lymph nodes after adequate lymphadenectomy.29 Patients with positive lymph nodes or those who had an inadequate lymphadenectomy were treated with EBRT on a standard pelvic field. Gastrointestinal symptoms (especially nausea, anorexia and ileus) occurred less frequently among patients treated with EBRT on a small pelvic field compared to those treated with standard pelvic field EBRT. These findings indicate a volume-based decrease of toxicity among high-risk EC patients treated with EBRT. EC patients without lymph node metastases after adequate surgical staging, in whom adjuvant EBRT is indicated, can be treated with EBRT limited to a small pelvic field, to reduce late adverse events.

Future perspectives

Optimizing standard treatment strategies The surgical staging system for EC was introduced by the International Federation of Gynecology and Obstetrics (FIGO) in 1988 and revised in 2009. For years, surgery by means of hysterectomy and BSO was exclusively considered safe when performed by an abdominal incision. After successful introduction of minimal invasive surgery for benign gynecological conditions, trials for applicability in cancer treatment were initiated. It is now generally recognized that surgery in patients with clinically stage I EC can be performed by laparoscopy, provided that this procedure is performed by an experienced gynecologic oncologist. 30•31

Although the FIGO recommended complete lymph node dissection for adequate surgical staging, the routine use of lymphadenectomy in addition to hysterectomy and BSO has not been accepted as standard treatment. It is now generally agreed that lymphadenectomy can be omitted in early stage EC.32•33 Whether lymphadenectomy should be performed in intermediate-high risk EC patients, and to what extend, is still subject of debate.34 Especially patients with high-risk EC have increased risk for lymph node metastases, distant metastases and death due to EC. In these patients, adequate surgical staging might be helpful to determine prognosis and adjuvant treatment strategies. In cancer types such as breast cancer and vulvar cancer, sentinel lymph node mapping is a standard procedure to predict the presence of lymph node metastases. A complete lymphadenectomy is only performed in case of a positive sentinel lymph node and consequently reducing treatment-related toxicity. Whether this technique might also apply for EC has to be determined.35

At present, EC patients with preoperatively determined high-risk factors (grade 3 endometrioid, non-endometrioid tumor types and/ or cervical invasion) are treated with TAH+BSO and pelvic- and para-aortic lymphadenectomy to obtain full staging. For high-risk EC patients, lymph node status has direct consequences for adjuvant treatment; patients that qualify for adjuvant EBRT and have negative lymph nodes based on complete lymphadenectomy29 can be treated with EBRT on a small pelvic field instead of a standard pelvic field, resulting in less gastrointestinal side-effects (see chapter 7). These results also indicate that it would be interesting to explore if high-risk EC patients can be safely treated with other radiotherapy techniques to further reduce radiother-

135

Chapter 8

136

apy-associated toxicity. For example, "intensity-modulated radiation therapy" (IMRT) might improve treatment of the postoperative bed or other areas at risk for recurrent disease while sparing adjacent normal tissues.36,37 In addition, VBT is now widely used as adjuvant treatment in EC patients with high-intermediate risk factors causing significant less side-effects while effective for local control.28•38 Whether VBT will sufficiently prevent recurrent disease among high-risk EC patients has to be determined. Although adjuvant radiotherapy prevents locoregional recurrences, vaginal relapses can generally be resolved by surgery and radiotherapy without compromising survival. In this respect, the PORTEC 4 study was designed and will commence in the near future. This multicenter trial aims to determine if EC patients with high-intermediate risk-features can be treated with lower dose VBT or even surgery alone, to prevent "overtreatment".

Distant metastases often account for recurrent disease in patients with high-risk EC.39

Although adjuvant radiotherapy reduces the risk of recurrent disease, overall survival is not influenced. Currently, three large randomized trials investigating the use of sequential chemotherapy and radiotherapy in high-risk EC patients are ongoing; the PORTEC-3 trial and two studies by the Gynecologic Oncology Group (GOG); the GOG#249 and GOG#258 trial. The PORTEC-3 trial compares EBRT with concurrent EBRT and chemotherapy followed by chemotherapy in high-risk EC patients (FICO 1-11 with high-risk features or FICO III). In the GOG#249 trial, EBRT is compared with chemotherapy followed by VBT in patients with high-intermediate risk stage 1-11 EC. The GOG#258 trial compares concurrent EBRT and chemotherapy followed by chemotherapy to chemotherapy alone in patients with FICO III-IV. The outcome of these trials might help to determine the role of chemotherapy in high-risk EC. Considering the fact that UCS has a similar carcinogenesis to EC, but with worse prognosis, sequential radio- and chemotherapy should be considered in patients with UCS.

Current treatment regimens for EC patients are based on clinicopathological characteristics. However, strategies refining the prediction of (lymph node) metastases and prognosis in individual patients might add to prevent under- and overtreatment. Moreover, identification of cell biological markers that add to risk assessment might also be used as targets for therapeutic intervention. In breast cancer for example, molecular markers such as HER-2/neu, estrogen and progesterone receptors are strongly related to prognosis and are used as targets endocrine therapy. Especially type I EC is a hormonesensitive malignancy and expression of estrogen and progesterone receptors has a positive influence on survival.40 In contrast to breast cancer, anti-hormonal treatment does not have a role in primary treatment of EC. Progestagens have a proven beneficial effect in metastatic disease only by slowing down tumor growth but mainly for EC with low tumor grade and positive progesterone receptor expression.34

Following other cancer types, efforts are being made to identify cell biological markers that accurately predict prognosis in EC.41 In this respect, characteristics of primary and metastatic tumor lesions might contribute to the knowledge of the development of metastases. This assumption is based on the hypothesis is that a molecular profile shared by the primary tumor and the metastatic lesion is most relevant for disease spread and targeting such by therapy. For EC, preliminary data for several molecular targeted


agents such as trastuzumab, temsirolimus, erlotinib, sorafenib and bevacizumab are under investigation with mixed results.34

Immunotherapy as novel treatment strategy In exploring new treatment strategies for cancer, immunotherapy is an interesting option. Cancer immunotherapy is based on potentiating the power and specificity of the immune system in order to eradicate cancer cells and prevent tumor outgrowth. There are several strategies aiming to strengthen the anti-tumor activity. By adoptive T-cell transfer, T-lymphocytes are expanded ex vivo and subsequently infused in the donor in order to establish a tumor-specific immune response. One of the advantages of this strategy is the fact that large numbers of T-lymphocytes can be selected in vitro for possible reactivity against tumor antigens. Trials have been performed in patients with advanced breast cancer and melanoma with promising results.42,43 However, the clinical effect of T-cell transfer is diminished by the suppressive action of Tregs and downregulation of MHC class I.44 Strategies that counteract these negative elements need to be developed. In addition to Tregs and MHC class I downregulation, several other agents are able to inhibit the function of CTLs and prevent an effective anti-tumor response. Interestingly, !DO-induced suppression of CTLs can be reversed by 1-MT. Promising pre-clinical results were obtained by combining 1-MT with chemotherapy, demonstrating a more potent anti-tumor effect compared to the use of both the agents alone as tested in mouse models.45A6 When aiming for blockage of !DO-induced immune suppression, identification of intra-cellular tryptophan transporters such as LATl or alternative tryptophan transporters might be an interesting option.21 These transporters can be upregulated in cancer tissue.20·47.48 Moreover, LATl upregulation was related to increased IDO activity, functioning as an immune escape mechanism in breast cancer cells.20 More research on the exact structure and function of tryptophan transporters in combination with IDO activity in the cancer micro-environment is warranted. Identification of small molecule inhibitors of these transporters expressed in cancer cells would be highly useful in preventing tryptophan depletion of T-lymphocytes and restoring an adequate tumorspecific immune response. Immunotherapeutic strategies by blocking inhibitory receptors such as the CTL-associated antigen 4 (CTLA4) are being developed. CTLA4 is exclusively expressed on T-cells and is an important negative regulator of the immune system. CTLA4 inhibits co-stimulatory signaling provided by antigen presenting cells (APC) to activate T-lymphocytes and CTLA4 enhances presence of Tregs. CTLA4 was the first immune checkpoint to be clinically targeted by antibodies. This resulted in a broad enhancement of immune responses dependent on T-helper cells. The use of anti-CTLA4 antibodies has caused significant improvements in survival of patients with metastatic melanoma. Currently, trials in other cancer types are ongoing.49

Another interesting pathway involved in immune suppression is programmed death-1 (PD-1) receptor ( expressed on T-cells) and its ligands PD-Ll / -L2. Overexpression of PD-Ll was correlated to poor prognosis in several tumor types. Phase I trials investigating blockage of this pathway have demonstrated promising clinical responses in patients with advanced cancer. Severe drug-related toxicity was observed in 9-14% of the patients.50·51

137

Chapter 8

138

The possible role of therapeutic vaccines as immunotherapy is based on the discovery that patients can harbor CDS+ and CD4+ T-cells specific for antigens expressed on their tumors. Vaccination trials are currently ongoing and primarily aim at patients with metastatic cancers. For example, a vaccination trial in advanced ovarian cancer using long peptides of TP53 (p53) showed that antigen-specific T-cells can be induced, although without clinical significant reduction of tumor mass.52 Other possible targets might be cancer testis antigens (such as NY-ESOl, MAGE-A3, MAGE-A4) which represent a cluster of antigens expressed in adult germ line tissue as well as in a wide range of tumors.53 Furthermore, WT-1 54

, Trop-2 55,56 and Sperm protein 17 57 have been shown to be highly expressed in EC but its use for targeted immunotherapy needs to be explored.

Combining standard therapy with immunotherapy So far, immunotherapy studies have displayed moderate clinical efficacy. In order to optimize treatment for EC, combining standard anti-cancer treatment (e.g. chemo- and radiotherapy) with targeted (immuno) therapy might be an option. In this way, the immunomodulatory qualities of radio- and chemotherapy are being exploited.

Radiotherapy is directly toxic to cancer cells by causing double-stranded DNA breaks. In addition, radiotherapy has an immunomodulatory role and is able to cause immunogenic death of cancer cells in several ways. Cancer cells are made more susceptible for recognition and killing by cytotoxic T-lymphocytes (CTL) by upregulation of intracellular peptide pools and cell-surface expression of tumor-associated antigens (TAA), Fas (apoptosis-inducing receptor), co-stimulatory molecule B7.1, and MHC class I. Furthermore, trafficking of effector T-lymphocytes to the tumor site is stimulated by induction of proiflammatory cytokines, chemokines and adhesion molecules which cause increased cell-cell contact and actual killing of the cancer cell. In addition, radiation-induced changes in the tumor microenvironment and vasculature facilitate homing of APCs and effector T-lymphocytes to the tumor site.58,59 Preclinical studies of radiotherapy in combination with immunotherapy are ongoing. Animal studies have demonstrated that radiation therapy can improve the efficacy of anti-tumor immunotherapy (e.g. by vaccination) and leading to reduction tumor mass.59-

61 If this effect also holds in a patient-related setting has to be investigated. Chemotherapeutic agents can also increase susceptibility of cancer cells for immunemediated attack by altering the expression of TAAs and MHC class 1.62 In addition, certain agents can induce an "immunogenic death" of cancer cells leading to activation of dendritic cells and subsequent antigen presentation and cross-presentation to T-cells. In this way, a more efficient CTL-mediated killing of cancer cells is attained.58 The role of chemotherapy in EC is currently not proven and randomized trials are currently ongoing. For chemotherapeutics to reach full efficacy, activation of immune mechanisms might be essential, and visa versa.63


Conclusions

During the last years, medical knowledge and treatment options for cancer are rapidly evolving. This has led to promising results, especially for patients with rare and complex malignancies. As a consequence, discussions were raised on how to organize health care in a way that optimal treatment for each patient is guaranteed. In the Netherlands, treatment of highly complex and/ or infrequent diseases has been allocated to specialized institutes where multidisciplinary teams are directly available with up to date knowledge. In the past, all patients with a certain cancer type were treated alike. New technologies and information on molecular markers with prognostic significance are now leading towards a more personalized approach. It is to be expected that, as in other cancer types, "patient-tailored" treatment is elaborated in EC and ensuring the best treatment for an individual patient with EC. For the future, we envision more personalized treatment strategies for patients with EC, using knowledge of specific subtypes as guide for treatment. In order to improve prognosis of patients with high-risk subtypes, future research should focus on new and combined therapies (e.g. radiotherapy, chemotherapy, targeted therapy and immunotherapy). For patients with good prognosis, current treatment modalities must be improved by new techniques to reduce side-effects as much as possible.

139

Chapter 8

140

References

1. Creutzberg CL, van Putten WL, Koper PC, Lybeert ML, Jobsen JJ, Warlam-Rodenhuis CC, et al. Surgery and postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomized trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet 2000;355:1404-1411.

2. Dunn GP, Old LJ, Schreiber RD. The three Es of cancer immunoediting. Annu Rev Immunol 2004;22:329-360.

3. Kondratiev S, Sabo E, Yakirevich E, Lavie 0, Resnick MB. Intratumoral CDS+ T lymphocytes as a prognostic factor of survival in endometrial carcinoma. Clin Cancer Res 2004;10:4450-4456.

4. Sato E, Olson SH, Ahn J, Bundy B, Nishikawa H, Qian F, et al. Intraepithelial CDS+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proc Natl Acad Sci U S A 2005;102:18538-18543.

5. Naito Y, Saito K, Shiiba K, Ohuchi A, Saigenji K, Nagura H, et al. CDS+ T cells infiltrated within cancer cell nests as a prognostic factor in human colorectal cancer. Cancer Res 1998;58:3491-3494.

6. Leffers N, Gooden MJ, de Jong RA, Hoogeboom BN, ten Hoor KA, Hollema H, et al. Prognostic significance of tumor-infiltrating T-lymphocytes in primary and metastatic lesions of advanced stage ovarian cancer. Cancer I mmunol Immunother 2009;58:449-459.

7. Bijen CB, Bantema-Joppe EJ, de Jong RA, Leffers N, Maurits MJ, Eggink HF, et al. The prognostic role of classical and nonclassical MHC class I expression in endometrial cancer. Int J Cancer 2010;126:1417-1427.

8. Kloor M, Becker C, Benner A, Woerner SM, Gebert J, Ferrone S, et al. Immunoselective pressure and human leukocyte antigen class I antigen machinery defects in microsatellite unstable colorectal cancers. Cancer Res 2005;65:6418-6424.

9. Dierssen JW, de Miranda NF, Ferrone S, van Puijenbroek M, Cornelisse CJ, Fleuren GJ, et al. HNPCC versus sporadic microsatellite-unstable colon cancers follow different routes toward loss of HLA class I expression. BMC Cancer 2007;7:33.

10. Linnebacher M, Gebert J, Rudy W, Woerner S, Yuan YP, Bork P, et al. Frameshift peptidederived T-cell epitopes: a source of novel tumor-specific antigens. Int J Cancer 2001;93:6-ll .

11 . Uyttenhove C, Pilotte L, Theate I, Stroobant V, Colau D, Parmentier N, et al . Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat Med 2003;9: 1269-1274.

12. Brandacher G, Perathoner A, Ladurner R, Schneeberger S, Obrist P, Winkler C, et al. Prognostic value of indoleamine 2,3-dioxygenase expression in colorectal cancer: effect on tumorinfiltrating T cells. Clin Cancer Res 2006;12:1144-1151.

13. Inaba T, Ino K, Kajiyama H, Shibata K, Yamamoto E, Kondo S, et al. Indoleamine 2,3-dioxygenase expression predicts impaired survival of invasive cervical cancer patients treated with radical hysterectomy. Gynecol Oncol 2010;117:423-428.

14. I naba T, Ino K, Kajiyama H, Yamamoto E, Shibata K, Nawa A, et al. Role of the immunosuppressive enzyme indoleamine 2,3-dioxygenase in the progression of ovarian carcinoma. Gynecol Oncol 2009;115:185-192.

15. Ino K, Yoshida N, Kajiyama H, Shibata K, Yamamoto E, Kidokoro K, et al. Indoleamine 2,3-dioxygenase is a novel prognostic indicator for endometrial cancer. Br J Cancer 2006;95:1555-1561.


16. Pan K, Wang H, Chen MS, Zhang HK, Weng DS, Zhou J, et al. Expression and prognosis role of indoleamine 2,3-dioxygenase in hepatocellular carcinoma. J Cancer Res Clin Oncol 2008;134:1247-1253.

17. Ino K, Yamamoto E, Shibata K, Kajiyama H, Yoshida N, Terauchi M, et al. Inverse correlation between tumoral indoleamine 2,3-dioxygenase expression and tumor-infiltrating lymphocytes in endornetrial cancer: its association with disease progression and survival. Clin Cancer Res 2008;14:2310-2317.

18. Munn DH, Sharma MD, Baban B, Harding HP, Zhang Y, Ron D, et al. GCN2 kinase in T cells mediates proliferative arrest and anergy induction in response to indoleamine 2,3-dioxygenase. Immunity 2005;22:633-642.

19. Kudo Y, Boyd CA. The role of L-tryptophan transport in L-tryptophan degradation by indoleamine 2,3-dioxygenase in human placental explants. J Physiol 2001;531:417-423.

20. Travers MT, Gow IF, Barber MC, Thomson J, Sherman DB. Indoleamine 2,3-dioxygenase activity and L-tryptophan transport in human breast cancer cells. Biochirn Biophys Acta 2004;1661:106-112.

21. Silk JD, Lakhal S, Laynes R, Vallius L, Karydis I, Marcea C, et al. IDO induces expression of a novel tryptophan transporter in mouse and human tumor cells. J Irnmunol 2011;187:1617-1625.


23. Kounelis S, Jones MW, Papadaki H, Bakker A, Swalsky P, Finkelstein SD. Carcinosarcomas (malignant mixed mullerian tumors) of the female genital tract: comparative molecular analysis of epithelial and mesenchymal components. Hum Pathol 1998;29:82-87.

24. Wada H, Enomoto T, Fujita M, Yoshino K, Nakashima R, Kurachi H, et al. Molecular evidence that most but not all carcinosarcomas of the uterus are combination tumors. Cancer Res 1997;57:5379-5385.

25. Amant F, Moerman P, Neven P, Timmerman D, Van Limbergen E, Vergote I. Endometrial cancer. Lancet 2005;366:491-505.

26. Silverberg SG, Major FJ, Blessing JA, Fetter B, Askin FB, Liao SY, et al. Carcinosarcorna (malignant mixed mesodermal tumor) of the uterus. A Gynecologic Oncology Group pathologic study of 203 cases. Int J Gynecol Pathol 1990;9:1-19.

27. Sreenan JJ, Hart WR. Carcinosarcomas of the female genital tract. A pathologic study of 29 metastatic tumors: further evidence for the dominant role of the epithelial component and the conversion theory of histogenesis. Am J Surg Pathol 1995;19:666-674.

28. Nout RA, Smit VT, Putter H, Jurgenliemk-Schulz IM, Jobsen JJ, Lutgens LC, et al. Vaginal brachytherapy versus pelvic external beam radiotherapy for patients with endometrial cancer of high-intermediate risk (PORTEC-2): an open-label, non-inferiority, randomised trial. Lancet 2010;375:816-823.

29. Nijman HW, Khalifa M, Covens A. What is the number of lymph nodes required for an "adequate" pelvic lymphadenectomy? Eur J Gynaecol Oncol 2004;25:87-89.

30. Mourits MJ, Bijen CB, Arts HJ, Ter Brugge HG, van der Sijde R, Paulsen L, et al. Safety of laparoscopy versus laparotorny in early-stage endometrial cancer: a randomised trial. Lancet Oncol. 2010;11:763-771.

31. Vergote I, Arnant F, Neven P. Laparoscopic hysterectomy for early endometrial cancer. Lancet Oncol. 2010;11 :707-708.

32. Benedetti Panici P, Basile S, Maneschi F, Alberto Lissoni A, Signorelli M, Scambia G, et al. Systematic pelvic lymphadenectomy vs. no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst 2008;100:1707-1716.

141

Chapter 8

142

33. Kitchener H, Swart AM, Qian Q, Amos C, Parmar MK. Efficacy of systematic pelvic lymphadenectomy in endometrial cancer (MRC ASTEC trial): a randomised study. Lancet 2009;373:125-136.

34. Wright JD, Barrena Medel NI , Sehouli J, Fujiwara K, Herzog TJ. Contemporary management of endometrial cancer. Lancet 2012;379:1352-1360.

35. Kang S, Yoo HJ, Hwang JH, Lim MC, Seo SS, Park SY. Sentinel lymph node biopsy in endometrial cancer: meta-analysis of 26 studies. Gynecol Oncol 2011;123:522-527.

36. Jhingran A, Winter K, Portelance L, Miller B, Salehpour M, Gaur R, et al. A Phase II Study of Intensity Modulated Radiation Therapy to the Pelvis for Postoperative Patients With Endometrial Carcinoma: Radiation Therapy Oncology Group Trial 0418. I nt J Radiat Oncol Biol Phys 2012 Apr 27.

37. Bouchard M, Nadeau S, Gingras L, Raymond PE, Beaulieu F, Beaulieu L, et al. Clinical outcome of adjuvant treatment of endometrial cancer using aperture-based intensity-modulated radiotherapy. Int J Radiat Oncol Biol Phys 2008;71:1343-1350.

38. Nout RA, Putter H, Jurgenliemk-Schulz IM, Jobsen JJ, Lutgens LC, van der Steen-Banasik EM, et al. Quality of life after pelvic radiotherapy or vaginal brachytherapy for endometrial cancer: first results of the randomized PORTEC-2 trial. J Clin Oncol 2009;27:3547-3556.

39. Creutzberg CL, van Putten WL, Warlam-Rodenhuis CC, van den Bergh AC, De Winter KA, Koper PC, et al. Outcome of high-risk stage I C, grade 3, compared with stage I endometrial carcinoma patients: the Postoperative Radiation Therapy in Endometrial Carcinoma Trial. J Clin Oncol 2004;22:1234-1241.

40. Jongen V, Briet J, de Jong R, ten Hoor K, Boezen M, van der Zee A, et al. Expression of estrogen receptor-alpha and -beta and progesterone receptor-A and -B in a large cohort of patients with endometrioid endometrial cancer. Gynecol Oncol 2009;112:537-542.

41. Nout RA, Bosse T, Creutzberg CL, Jurgenliemk-Schulz IM, Jobsen JJ, Lutgens LC, et al. Improved risk assessment of endometrial cancer by combined analysis of MSI, PI3K-AKT, Wnt/ beta-catenin and P53 pathway activation. Gynecol Oncol 2012 May 15.

42. Hardy NM, Mossoba ME, Steinberg SM, Fellowes V , Yan XY, Hakim FT, et al. Phase I trial of adoptive cell transfer with mixed-profile type-I /type-II allogeneic T cells for metastatic breast cancer. Clin Cancer Res 2011;17:6878-6887.

43. Rosenberg SA, Yang JC, Sherry RM, Kammula US, Hughes MS, Phan GQ, et al. Durable complete responses in heavily pretreated patients with metastatic melanoma using T-cell transfer immunotherapy. Clin Cancer Res 2011 Jul 1;17(13):4550-4557.

44. Yao X, Ahmadzadeh M, Lu YC, Liewehr DJ, Dudley ME, Liu F, et al. Levels of peripheral CD4+FoxP3+ regulatory T cells are negatively associated with clinical response to adoptive immunotherapy of human cancer. Blood 2012 May 3.

45. Hou DY, Muller AJ, Sharma MD, DuHadaway J, Banerjee T, Johnson M, et al. Inhibition of indoleamine 2,3-dioxygenase in dendritic cells by stereoisomers of 1-methyl-tryptophan correlates with antitumor responses. Cancer Res 2007;67:792-801.

46. Qian F, Villella J, Wallace PK, Mhawech-Fauceglia P, Tario JD,Jr., Andrews C, et al. Efficacy of levo-1-methyl tryptophan and dextro-1-methyl tryptophan in reversing indoleamine-2,3-dioxygenase-mediated arrest of T-cell proliferation in human epithelial ovarian cancer. Cancer Res 2009;69:5498-5504.

47. Fan X, Ross DD, Arakawa H, Ganapathy V, Tamai I, Nakanishi T. I mpact of system L amino acid transporter 1 (LATl) on proliferation of human ovarian cancer cells: a possible target for combination therapy with anti-proliferative aminopeptidase inhibitors. Biochem Pharmacol 2010;80:811-818.


48. Yanagida 0, Kanai Y, Chairoungdua A, Kim DK, Segawa H, Nii T, et al. Human L-type amino acid transporter 1 (LATl): characterization of function and expression in tumor cell lines. Biochim Biophys Acta 2001;1514:291-302.

49. Pardall OM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer 2012;12:252-264.

50. Brahmer JR, Tykodi SS, Chow LQ, Hwu WJ, Topalian SL, Hwu P, et al. Safety and Activity of Anti-PD-LI Antibody in Patients with Advanced Cancer. N Engl J Med 2012 Jun 2.

51. Topalian SL, Hodi FS, Brahmer JR, Gettinger SN, Smith DC, McDermott OF, et al. Safety, Activity, and Immune Correlates of Anti-PD-1 Antibody in Cancer. N Engl J Med 2012 Jun 2.

52. Leffers N, Vermeij R, Hoogeboom BN, Schulze UR, Wolf R, Hamming IE, et al. Long-term clinical and immunological effects of p53-SLP(R) vaccine in patients with ovarian cancer. Int J Cancer 2012;130:105-112.

53. Brooks N, Pouniotis OS. Immunomodulation in endometrial cancer. Int J Gynecol Cancer 2009;19:734-740.

54. Coosemans A, Wolfl M, Berneman ZN, Van Tendeloo V, Vergote I, Amant F, et al. Immunological response after therapeutic vaccination with WTl mRNA-loaded dendritic cells in endstage endometrial carcinoma. Anticancer Res 2010;30:3709-3714.

55. Varughese J, Cocco E, Bellone S, de Leon M, Bellone M, Todeschini P, et al. Uterine serous papillary carcinomas overexpress human trophoblast-cell-surface marker (trop-2) and are highly sensitive to immunotherapy with hRS7, a humanized anti-trop-2 monoclonal antibody. Cancer 2011;117:3162-72.

56. Bignotti E, Ravaggi A, Romani C, Palchetti M, Lonardi S, Facchetti F, et al. Trop-2 overexpression in poorly differentiated endometrial endometrioid carcinoma: implications for immunotherapy with hRS7, a humanized anti-trop-2 monoclonal antibody. Int J Gynecol Cancer 2011;21:1613-1621.

57. Li FQ, Liu Q, Han YL, Wu B, Yin HL. Sperm protein 17 is highly expressed in endometrial and cervical cancers. BMC Cancer 2010;10:429.

58. Hodge JW, Ardiani A, Farsaci B, Kwilas AR, Gameiro SR. The tipping point for combination therapy: cancer vaccines with radiation, chemotherapy, or targeted small molecule inhibitors. Semin Oneal 2012;39:323-339.

59. Demaria S, Bhardwaj N, McBride WH, Formenti SC. Combining radiotherapy and immunotherapy: a revived partnership. Int J Radiat Oneal Biol Phys 2005;63:655-666.

60. Rei ts EA, Hodge JW, Herberts CA, Groothuis TA, Chakraborty M, Wansley EK, et al. Radiation modulates the peptide repertoire, enhances MHC class I expression, and induces successful antitumor immunotherapy. J Exp Med 2006;203:1259-1271.

61 . Meng Y, Efimova EV, Hamzeh KW, Darga TE, Mauceri HJ, Fu YX, et al. Radiation-inducible Immunotherapy for Cancer: Senescent Tumor Cells as a Cancer Vaccine. Mal Ther 2012 May;20:1046-1055.

62. Bao L, Dunham K, Lucas K. MAGE-Al, MAGE-A3, and NY-ESO-1 can be upregulated on neuroblastoma cells to facilitate cytotoxic T lymphocyte-mediated tumor cell killing. Cancer Immunol Immunother 2011;60:1299-1307.

63. Kandalaft LE, Singh N, Liao JB, Facciabene A, Berek JS, Powell DJ,Jr, et al. The emergence of immunomodulation: combinatorial immunochemotherapy opportunities for the next decade. Gynecol Oncol 2010;116:222-233.

143

Chapter 9

Summary in Dutch

(Nederlandse samenvatting)

Summan; in Dutch (Nederlandse samenvatting)

Inleiding

Endometriumcarcinoom In Nederland wordt jaarlijks bij circa 1900 vrouwen endometriumcarcinoom (baarmoederkanker) vastgesteld. Endometriumcarcinoom is de meest voorkomende gynaecologische maligniteit en de vierde meest voorkomende maligniteit bij vrouwen in de westerse wereld, na borstkanker, longkanker en darmkanker. Chirurgie vormt de basis voor de behandeling van endometriumcarcinoom en bestaat uit het verwijderen van de uterus en eierstokken. Wanneer voorafgaand aan de operatie wordt vastgesteld dat het een endometriumcarcinoom met een hoog risicoprofiel betreft, worden ook de lymfeklieren in het bekken verwijderd (lymfadenectomie). Na de operatie wordt bepaald of er een aanvullende behandeling nodig is. Bij patienten met een laag risicoprofiel is er geen aanvullende behandeling noodzakelijk omdat zij een laag risico hebben op het ontwikkelen van een tumorrecidief. Patienten met een laag-intermediair risicoprofiel worden wel aanvullend behandeld door middel van inwendige bestraling (brachytherapie). Patienten met hoog risicoprofiel en/ of positieve lymfeklieren krijgen uitwendige radiotherapie. Op dit moment is chemotherapie nog geen onderdeel van de standaardbehandeling van endometriumcarcinoom, maar zij kan worden overwogen bij patienten met een gevorderd stadium van de ziekte (stadium III of IV). Wetenschappelijk onderzoek zal moeten uitwijzen of chemotherapie een plaats zal kunnen krijgen in de behandeling en welke groep patienten hier het meeste bij gebaat zal zijn. Over het algemeen hebben patienten met endometriumcarcinoom een goede prognose, doordat de ziekte meestal in een vroeg stadium wordt ontdekt. Er zijn echter een paar agressieve subtypen (clearcell carcinoom, sereus carcinoom en carcinosarcoom) waarbij de overlevingscijfers dramatisch dalen. Vooral in deze specifieke gevallen is het van groot belang dat er nieuwe en betere behandelingsmogelijkheden worden gevonden om daarmee de prognose te verbeteren.

De rol van het immuunsysteem in kanker Het immuunsysteem speelt een belangrijke rol in de ontwikkeling van kanker. Immunotherapie zou daarom een nieuwe behandelmethode kunnen zijn voor endometriumcarcinoom. Immunotherapie is erop gericht de eigenschappen van het immuunsysteem te versterken om tumorcellen te vernietigen. Het immuunsysteem bestaat uit twee componenten; het aangeboren en verworven immuunsysteem. Het aangeboren immuunsysteem, ook wel het niet-specifieke immuunsysteem, zorgt met name voor de eerstelijnsafweer van het lichaam. Het verworven immuunsysteem, ook wel het specifieke immuunsysteem, heeft tijd nodig zich te ontwikkelen maar kan daarna een zeer effectieve en specifieke afweerreactie veroorzaken. In dit proefschrift wordt alleen ingegaan op het verworven immuunsysteem omdat deze component betrokken is bij de specifieke herkenning van en aanval op kankercellen. Een centrale rol in de strijd tegen kankercellen wordt vervuld door T-cellen (ook wel: T-lymfocyten). Cytotoxische T-cellen zijn in staat kankercellen te herkennen als "lichaamsvreemd" en kunnen deze vervolgens vernietigen. Echter, cytotoxische T-cellen kunnen kankercellen pas herkennen wanneer tumorspecifieke antigenen tot expressie

147

Chapter 9

148

warden gebracht op het celoppervlak van de kankercel. Deze antigenen moeten op hun beurt gepresenteerd warden door MHC (major histocompatibility complex) moleculen. Geheugen-T-cellen zijn in staat een immuunreactie te herinneren, gericht op een specifiek antigen. Wanneer in de toekomst opnieuw een invasie van (kanker)cellen met het bekende/ geregistreerde antigen plaatsvindt, zal de immuunreactie veel sneller, en dus efficienter, kunnen plaatsvinden. In tegenstelling tot de eerder genoemde T-cellen, onderdrukken regulatoire T-cellen juist de werking van cytotoxische T-cellen. Hiermee wordt voorkomen dat het immuunsysteem "overactief" reageert, bijvoorbeeld in het geval van een auto-immuunreactie. Het immuunsysteem is niet altijd voldoende effectief om de groei van kankercellen te remmen of te vernietigen. Dit blijkt uit het feit dat kankercellen zich nog steeds ontwikkelen tot een klinisch detecteerbare tumorrnassa. Kankercellen hebben eigenschappen ontwikkeld waardoor ze kunnen ontsnappen aan het irnmuunsysteem; bijvoorbeeld door de aanwezigheid van regulatoire T-cellen te verhogen zodat er een "tolerante" status ten opzichte van de "lichaamsvreernde" kankercellen ontstaat. Daarnaast leidt afschakeling van het MHC complex ertoe dat cytotoxische T-cellen kankercellen niet meer kunnen herkennen. Als gevolg hiervan kan er ook geen vernietiging van deze cellen plaatsvinden. Een andere manier om de functie van T-cellen te verminderen is de productie van indoleamine 2,3-dioxygenase (IDO). Dit enzym katalyseert de afbraak van het essentiele aminozuur tryptofaan. Net als iedere eel in het lichaarn hebben T-cellen tryptofaan nodig om te kunnen overleven. Doordat turnoren grotere hoeveelheden IDO produceren dan gezonde weefsels wordt de beschikbaarheid van tryptofaan dusdanig verlaagd dat T-cellen minder effectief warden of uiteindelijk doodgaan.

Summary in Dutch (Nederlandse samenvatting)

Dit proefschrift

Dit proefschrift gaat in op de interactie tussen het immuunsysteem en endometriumcarcinoom. Daarnaast wordt aandacht besteed aan het biologisch gedrag van subtypen van endometriumcarcinoom met een hoog risicoprofiel. Ook worden bijwerkingen geevalueerd als gevolg van de behandeling van endometriumcarcinoom met een hoog risicoprofiel. Deze verworven kennis moet uiteindelijk bijdragen aan de verbetering van de behandeling van patienten met endometriumcarcinoom.

Interactie tussen immuunsysteem en endometriumcarcinoom Voordat immunotherapie kan worden ingevoerd als behandeling van endometriumcarcinoom is meer kennis over de interactie tussen immuunsysteem en endometriumcarcinoom noodzakelijk. In hoofdstuk 2 werd onderzocht of T-cellen invloed hebben op de prognose van patienten met endometriumcarcinoom. Tumorweefsel van 368 patienten werd geevalueerd en in het merendeel van deze tumoren waren cytotoxische T-cellen, geheugen-T-cellen en regulatoire T-cellen aanwezig. Dit impliceert dat tumorspecifieke immuniteit aanwezig is in endometriumcarcinoom. De aanwezigheid van veel tumorinfiltrerende cytotoxische T-cellen en geheugen-T-cellen voorspelde een betere overleving voor patienten. Daarnaast bleek de verhouding tussen cytotoxische T-cellen en regulatoire T-cellen een goede voorspeller voor de overleving van patienten met endometriumcarcinoom. Deze bevindingen zijn een belangrijke ondersteuning voor de ontwikkeling van immunotherapie voor patienten met endometriumcarcinoom.

Voor een efficiente toepassing van immunotherapie is kennis van mechanismen waardoor kankercellen ontsnappen aan het immuunsysteem onontbeerlijk. Afschakeling van MHC klasse I is een bekend ontsnappingsmechanisme. Onderzoek heeft aangetoond dat patienten met darmkanker, bij wie het normale DNA-reparatiesysteem niet goed werkt, een sterkere immuunreactie kunnen ontwikkelen tegen kankercellen. Dit komt doordat een defect DNA-reparatiesysteem ervoor zorgt dat er meer antigenen tot expressie komen op de kankercellen. In theorie zou dit kunnen leiden tot een betere prognose voor deze patientengroep. Opvallend genoeg is juist bij deze tumoren expressie van het MHC-molecuul vaker afgeschakeld, waardoor ze kunnen ontsnappen aan een immuunreactie. De resultaten van het onderzoek beschreven in hoofdstuk 3 toonden dat dit mechanisme ook geldt voor patienten met een niet-erfelijk type endometriumcarcinoom; er werd een verband gevonden tussen afwijkende expressie van belangrijke DNAreparatie-eiwitten en afgeschakeld MHC klasse I. Daarnaast bleek dat patienten met een afwijkende expressie van DNA-reparatie-eiwitten een slechtere prognose hebben dan patienten met een intacte expressie. Waarschijnlijk kan dit verklaard worden doordat in de eerste groep patienten ook het MHC klasse I vaker "uit" staat waardoor de cytotoxische T-cellen de kankercellen niet kunnen herkennen en niet kunnen doden. In dit hoofdstuk werd ook duidelijk dat een hoog aantal tumorinfiltrerende cytotoxische T-cellen een positief effect heeft op de overleving van patienten met een normale expressie van DNA-reparatie-eiwitten. Het aantal cytotoxische T-cellen heeft echter geen invloed op de prognose in patienten met afwijkende expressie van DNA-reparatieeiwitten.

149

Chapter 9

150

Immunotherapie die erop gericht is het aantal tumorspecifieke cytotoxische T-cellen te verhogen zal zeer waarschijnlijk minder effectief zijn in patienten met een afgeschakeld MHC klasse I en/ of afwijkend DNA-reparatiesysteem. Deze groep kan mogelijk geselecteerd worden door de status van MHC en DNA-reparatie-eiwitten te bepalen in tumormateriaal verkregen na chirurgie. Deze specifieke groep heeft mogelijk baat bij therapeutische strategieen die MHC klasse I expressie induceren.

IDO is een antler mechanisme waardoor kankercellen kunnen ontsnappen aan een immuunreactie. In hoofdstuk 4 werd de IDO-activiteit gemeten in serum van patienten met endometrium-, ovarium- en vulvacarcinoom. Hiertoe werden concentraties van tryptofaan en kynurenine (eerste afbraakproduct van tryptofaan) bepaald. De kynurenine-/tryptofaan-ratio staat hierbij ongeveer gelijk aan de IDO activiteit. De kynurenine-/ tryptofaan-ratio was significant verhoogd in serum van patienten met een actieve maligniteit, vergeleken met serum van gezonde vrouwen. Deze resultaten impliceren dat patienten met kanker een hogere systemische IDO-activiteit hebben. Studies in meerdere tumortypen hebben aangetoond dat expressie van IDO in tumorweefsel gepaard gaat met een slechtere prognose. Het negatieve effect van IDO wordt veroorzaakt door verhoogde afbraak van tryptofaan. Daling van tryptofaan leidt tot een verminderde functie en dood van tumorspecifieke T-cellen.

In hoofdstuk 5 werd getoond dat in tumorweefsel met hoge IDO-expressie een kleiner aantal tumorinfiltrerende cytotoxische T-cellen aanwezig is dan in tumorweefsel met weinig/ geen IDO-expressie. Daarnaast bleek dat hoge IDO-expressie gerelateerd is aan een slechtere prognose voor patienten met endometriumcarcinoom. Mogelijk kunnen deze patienten baat hebben bij behandelingen waarbij IDO-activiteit wordt geblokkeerd. Preklinische modellen hebben aangetoond dat IDO kan warden onderdrukt door 1-methyltryptofaan (1-MT). Voornamelijk de combinatie van 1-MT met chemotherapie bleek te zorgen voor herstel van functie en stijging van het aantal T-cellen. Daarnaast zorgde deze combinatie voor een afname van tumormassa in een muismodel. Klinisch onderzoek zal moeten uitwijzen of dit ook op menselijke patienten met kanker van toepassing is.

Het onderzoek beschreven in hoofdstuk 5 ging verder in op het immunosuppressieve mechanisme van IDO; hoe is het mogelijk dat T-cellen last hebben van een tekort aan tryptofaan in tegenstelling tot kankercellen, terwijl ze in dezelf de micro-omgeving leven? Allereerst werd de specifieke lokalisatie van IDO in het cytoplasma onderzocht door gebruik te maken van elektronenmicroscopie. Kankercellijnen werden behandeld met verschillende concentraties IFN-y (natuurlijke inducerende stof voor IDO). Zoals verwacht nam IDO expressie in de kankercellen toe naarmate ze behandeld waren met hogere concentraties IFN-y. Daarnaast werd aangetoond dat de cytoplasmatische expressie van IDO voornamelijk net onder het membraan van de kankercel gelokaliseerd is. Door middel van massaspectometrie werd aangetoond dat de tryptofaandaling en kynureninestijging (als gevolg van IDO) voornamelijk plaatsvindt in het extracellulaire milieu van de kankercel. Deze veranderingen vinden in veel mindere mate ook intracellulair plaats. Blijkbaar is de dramatische daling van tryptofaan funest voor T-cellen,


terwijl de intracellulaire tryptofaanconcentratie net op een niveau blijft dat tumorcellen kunnen overleven. De subcellulaire locatie van IDO is een mogelijke verklaring voor dit mechanisme. Een andere mogelijkheid is dat kankercellen bepaalde tryptofaantransporters tot expressie brengen. Hierdoor wordt de intracellulaire concentratie van tryptofaan op een functioneel niveau gehandhaafd, ten koste van de extracellulaire concentratie. ldealiter zou therapie specifiek gericht op blokkade van deze transporters kunnen zorgen voor herstel van de functie en stijging van het aantal tumorspecifieke T-cellen.

Carcinosarcoom; agressief subtype van endometriumcarcinoom Het carcinosarcoom is een zeldzaam maar zeer agressief subtype van het endometriumcarcinoom. Deze tumor bestaat uit twee kwaadaardige componenten; een sarcoomen carcinoomcomponent. De sarcoomcomponent ontstaat door extreme differentiatie vanuit de carcinoomcomponent. Tot een aantal jaren geleden werd het carcinosarcoom beschouwd als een sarcoom en ook op die manier behandeld. Volgens de huidige richtlijnen wordt een carcinosarcoom van de uterus op dezelfde manier behandeld als een hooggradig endometriumcarcinoom. Echter, de prognose van een patient met een carcinosarcoom is beduidend slechter dan een hooggradig endometriumcarcinoom, zoals hoofdstuk 6 laat zien. Dit hoofdstuk beschrijft een cohort van 40 patienten met een carcinosarcoom van de uterus. Om meer te weten te komen over eigenschappen en gedrag van deze tumorsoort werd een groot aantal moleculaire markers onderzocht in het tumorweefsel van deze patienten. Hieruit bleek dat er grote overeenkomsten bestaan tussen carcinosarcoom en endometriumcarcinoom. Deze resultaten impliceren dat beide tumoren op eenzelfde manier ontstaan. Van beide componenten is de carcinoomcomponent de "drijvende kracht"; de meerderheid van de metastasen (72%) en vaatinvasie (70%) werd alleen veroorzaakt door de carcinoomcomponent. Daarnaast bleken histologische eigenschappen van de carcinoomcomponent van invloed te zijn op de prognose, net als bij het endometriumcarcinoom. Al deze bevindingen dragen bij aan de kennis van het carcinosarcoom. Toekomstige behandelingen zullen zich vooral moeten richten op eigenschappen van de carcinoomcomponent.

Behande lingsgere lateerde toxici tei t Chirurgie vormt de basis van de behandeling van endometriumcarcinoom. Adjuvante radiotherapie is de meest gebruikte aanvullende behandeling en kan zowel inwendig als uitwendig als door een combinatie van beide worden gegeven. Radiotherapie kan gepaard gaan met bijwerkingen op korte en lange termijn. Daarom is het van belang dat alleen patienten worden behandeld die daadwerkelijk gebaat zijn bij deze therapie. Veel onderzoek wordt verricht om de behandelingen dusdanig te verbeteren dat de effectiviteit wordt verhoogd en de bijwerkingen worden gereduceerd. Patienten met een intermediair risicoprofiel worden alleen nog aanvullend behandeld door inwendige bestraling. Vergeleken met uitwendige radiotherapie veroorzaakt inwendige radiotherapie minder bijwerkingen terwijl de effectiviteit om recidieven te voorkomen gelijk blijft. Uitwendige bestraling is daarom alleen nog gemdiceerd bij patienten met een hoog risicoprofiel. In hoofdstuk 7 werd de behandelingsgerelateerde toxiciteit geevalueerd onder 75 patienten met een endometriumcarcinoom met een hoog risicoprofiel. Deze patienten

151

Chapter 9

152

werden behandeld met chirurgie (verwijdering van uterus, eierstokken en lymfadenectomie) en adjuvante uitwendige radiotherapie. Patienten met negatieve lymfeklieren op basis van een complete lymfadenectomie werden bestraald op een beperkt bekkenveld (doelgebied: proximale helft vagina en parametria met als bovengrens S1-S2). Patienten met positieve pelviene lymfeklieren en/ of een inadequate lymfadenectomie werden bestraald op een standaard bekkenveld ( doelgebied: proximale helft vagina, parametria en pelviene lymfklierregio ). In de laatstgenoemde groep kwamen gastrointestinale bijwerkingen (misselijkheid, verminderde eetlust en ileus) significant vaker voor dan bij patienten die op een beperkt bekkenveld waren bestraald. Deze bevindingen pleiten voor een complete stadiering bij patienten met een hooggradig endometriumcarcinoom. Wanneer blijkt dat patienten een hoog risicoprofiel hebben en uitwendige radiotherapie gemdiceerd is, kan in geval van negatieve lymfeklieren bestraald worden op een beperkt bekkenveld, uiteindelijk resulterend in minder gastro-intestinale bijwerkingen. Daarnaast zal toekomstig onderzoek moeten uitwijzen of deze groep patienten kunnen worden bestraald met nog kleinere volumina (brachytherapy of intensiteits gemoduleerde radiotherapie (IMRT)) om zo de behandelingsgerelateerde toxiciteit nog verder terug te dringen.


Conclusie en toekomstperspectief

Hoewel endometriumcarcinoom de laagste mortaliteit heeft van alle gynaecologische maligniteiten, is verbetering van behandelingen en prognose noodzakelijk. Huidige behandelingsstrategieen kunnen geoptimaliseerd warden zodat ze enerzijds effectiever warden en anderzijds minder bijwerkingen veroorzaken. Wat betreft de chirurgische behandeling is er nag steeds geen consensus over de rol van een complete lymfadenectomie als onderdeel van de chirurgische stadiering. Analoog aan borst- en vulvacarcinoom zou de invoering van de schildwachtklierprocedure bij het endometriumcarcinoom kunnen leiden tot minder morbiditeit als gevolg van "overbodige" lymfadenectomie. Onderzoek naar mogelijke invoering van deze techniek staat echter nag in de kinderschoenen. Radiotherapie is de meest gebruikte aanvullende behandeling voor endometriumcarcinoom en gaat gepaard met significante bijwerkingen. Onderzoek naar nieuwe technieken om de doelvolumina en daarmee de bijwerkingen zo veel mogelijk te reduceren, is voortdurend gaande. Alleen patienten met een endometriumcarcinoom met een hoog risicoprofiel warden op dit moment nag uitwendig bestraald. Of vaginale brachytherapie of IMRT voldoende effectief is ter voorkoming van tumorrecidieven, zal moeten warden onderzocht. Een lokaal tumorrecidief kan over het algemeen effectief met chirurgie en/ of radiotherapie warden behandeld. Patienten met endometriumcarcinoom met een hoog risicoprofiel hebben echter grate kans op metastasen op afstand. Op dit moment wordt onderzocht of chemotherapie in combinatie met radiotherapie kan bijdragen aan een betere prognose voor patienten met endometriumcarcinoom. Nieuwe behandelvormen zoals immunotherapie warden op dit moment onderzocht voor veel kankersoorten. Zoals beschreven in dit proefschrift heeft het immuunsysteem een belangrijke rol in endometriumcarcinoom. Om de prognose van patienten met endometriumcarcinoom te verbeteren is immunotherapie een interessante optie. Hoewel bepaalde strategieen hebben kunnen leiden tot een significante stijging van de tumorspecifieke afweer is er vooralsnog geen effect op de overleving van patienten met kanker aangetoond. De oorzaak hiervan moet waarschijnlijk worden gezocht in het bestaan van de ontsnappingsmechanismen. Kennis van deze mechanismen is essentieel voordat ze uiteindelijk kunnen warden beinvloed om een effectieve immunotherapeutische strategie te ontwikkelen. Zeer waarschijnlijk zal een combinatie van immunotherapie met radio- en/ of chemotherapie effectiever zijn dan het toepassen van de behandelingen afzonderlijk. Onderzoek zal moeten uitwijzen of de prognose van patienten met endometriumcarcinoom inderdaad op deze manier kan worden verbeterd.

153

Abbreviations

Abbreviations ----- -----------

Abbreviations used:

156

1-MT

APC

l\m

BMI

BSO

CI

EC

ER

CTCAE

CTL

CTLA4

DFS

DSS

EBRT

EF

FIGO

GARpo

GOG

Gy

HEK

HLA

HR

IDO

IFN-y

IMRT

IQR

Kyn

LATl

LVSI

MCL

MHC

MLHl

MMR

MSH2

MSH6

MSI

MSS

1-methyltryptophan

antigen presenting cell

beta-2-microglobulin

body mass index

bilateral salpingo-oophorectomy

confidence interval


estrogen receptor

Common Terminology Criteria for Adverse Events

cytotoxic T-lymphocyte

CTL-associa ted antigen 4

disease-free survival

disease specific survival

external beam radiation therapy

extended field

International Federation of Gynecology and Obstetrics

goat anti mouse peroxidase-labeled

Gynecologic Oncology Group

gray

human embryonic kidney

human leukocyte antigen

hazard ratio

indoleamine 2,3-dioxygenase

interferon-gamma

intensity-modulated radiation therapy

interquartile range

kynurenine

system L amino acid transporter 1

lymphvascular space involvement

Medical center Leeuwarden

major histocompatibility complex

mutL homolog 1

mismatch repair

mutS homolog 2

mutS homolog 6

microsatellite instability

microsatellite stability

NA NAD NK cell NS OR OS PD-1 PD-L PR PTEN

SmPF StPF TAA TAH TCR TDO TIL TMA Treg Trp ucs

UMCG VBT

WHO XLC-MS/MS

not applicable nicotinamide adenine dinucleotide natural killer cell not significant odds ratio overall survival programmed death-1 programmed death-ligand progesterone receptor phosphatase and tensin homolog rabbit anti mouse peroxidase-labeled small pelvic field standard pelvic field tumor associated antigen total abdominal hysterectomy T-cell receptor tryptophan 2,3-dioxygenase tumor-infiltrating lymphocytes tissue microarray regulatory T-lymphocyte tryptophan uterine carcinosarcoma University Medical Center Groningen vaginal brachytherapy World Health Organization automated on-line solid-phase extraction-liquid chromatographic-tandem mass spectrometry

Abbreviations

157

List of co-authors

List of co-authors

160

Name and affiliation

Jan G. Aalders, MD, PhD Department of Gynecologic Oncology University Medical Center Groningen

Henriette J.G. Arts, MD, PhD Department of Gynecologic Oncology University Medical Center Groningen

Annemarie Boerma Department of Gynecologic Oncology University Medical Center Groningen

H. Marike Boezen, PhD Department of Epidemiology University Medical Center Groningen

Marloes J.M. Gooden, MD Department of Gynecologic Oncology University Medical Center Groningen

Harry Hollema, MD, PhD Department of Pathology University Medical Center Groningen

Klaske A. ten Hoor Department of Gynecologic Oncology University Medical Center Groningen

Ido P. Kema, PhD Department of Laboratory Medicine University Medical Center Groningen

Jasper Krijnen Department of Laboratory Medicine University Medical Center Groningen

Ninke Leffers, MD, PhD Department of Gynecologic Oncology University Medical Center Groningen

Marian J.E. Mourits, MD, PhD Department of Gynecologic Oncology University Medical Center Groningen

Hans W. Nijman, MD, PhD Department of Gynecologic Oncology University Medical Center Groningen

Elisabeth Pras, MD, PhD Department of Radiation Oncology University Medical Center Groningen

Anna K.L. Reyners, MD, PhD Department of Medical Oncology University Medical Center Groningen

Annerie Slot, MD Radiotherapy Institute Friesland

Paul R. Timmer, MD Department of Radiation Oncology Isala Clinics

Marcel Volmer, PhD Department of Laboratory Medicine University Medical Center Groningen

Johannes J.L. van der Want, PhD Molecular Imaging and Electron Microscopy Department of Cell Biology University Medical Center Groningen

Tera F. Wijbrandi, MD Department of Gynecologic Oncology University Medical Center Groningen

Ate G.J. van der Zee, MD, PhD Department of Gynecologic Oncology University Medical Center Groningen

List of co-authors

161

Dankwoord

164

Dankwoord

Dankwoord

Promoveren is . .. net zolang blijven zitten totdat het er staat.

Toen ik in mijn vierde studiejaar met mijn wetenschappelijke stage begon, had ik niet kunnen vermoeden <lat ik uiteindelijk een proefschrift zou schrijven. Nu, ruim 6 jaar later, is het zover! Hoewel een promotietraject af en toe een (flink) beroep doet op eigen wilskracht en doorzettingsvermogen, zou <lit proefschrift er niet gekomen zijn zonder de steun van velen. Daarom, en om niet het risico te lopen iemand te vergeten, wil ik iedereen die op enige manier heeft bijgedragen aan de totstandkoming van dit proefschrift heel hartelijk bedanken!

Toch wil ik hier een aantal personen in het bijzonder noemen.

Allereerst, prof. dr. H.W. Nijman, beste Hans, ik wil je ervoor bedanken dat je me ruim zes jaar geleden hebt betrokken bij het promotieonderzoek van Vincent Jongen. Sindsdien is mijn interesse in onderzoek alleen maar toegenomen. Dit is voor een groot gedeelte te danken aan de bevlogenheid en het aanstekelijke enthousiasme waarmee jij wetenschappelijk onderzoek doet. Je combineert schijnbaar moeiteloos wetenschappelijk onderzoek met bestuurlijke en klinische werkzaamheden. Jouw toegankelijkheid, manier van communiceren en verrassende humor zorgen ervoor <lat ik met een goed gevoel terugkijk op deze afgelopen periode.

Prof. dr. H. Hollema, beste Harry, jouw kennis, enthousiasme voor je vak en uitstekende didactische vaardigheden hebben ertoe geleid <lat ik echte interesse heh gekregen voor de pathologie. Gedurende vele onderwijsmomenten achter de microscoop konden we op een mooie manier discussieren over medische en niet-medische onderwerpen. De bij jou opgedane kennis helpt me om beter (klinisch) te denken.

Prof. dr. I.P. Kema, beste Ido, hoeveel hoogleraren kunnen zeggen dat het enzym <lat zij onderzoeken hun (voor)naam draagt ... ? Ik wil je bedanken voor het geduld waarmee je mijn kennis van (onder andere) de massaspectrometer probeerde bij te spijkeren. Er ging een heel andere wereld voor me open, ook doordat jij vaak de (onderzoeks)materie net even anders weet te belichten.

Prof. dr. H.M. Boezen, beste Marike, vooraf had ik eerlijk gezegd niet gedacht <lat ik statistiek zo'n leuk en interessant onderdeel van mijn promotieonderzoek zou gaan vinden. Jouw revisies van mijn manuscripten vond ik erg waardevol en verhelderend.

Dr. E. Pras, beste Betty, met jou kwam ik in contact tijdens de voorbereiding van hoofdstuk 7. Bedankt voor je enthousiaste en betrokken begeleiding tijdens het schrijven van <lit hoofdstuk en in de periode daarna. 1k bewonder de manier waarop jij jouw vak uitoefent. Je bezit ongelooflijk veel kennis van de radiotherapie. Jouw empathische houding richting patienten en collega's is een voorbeeld voor mij.

165

Dankwoord

166

Vele co-auteurs hebben bijgedragen aan de artikelen in dit proefschrift. Ik wil iedereen heel hartelijk bedanken voor hun onmisbare medewerking.

De leden van de leescommissie, prof. dr. F. Amant, prof. dr. J.A. Langendijk en prof. dr. L.F.M.H. de Leij ben ik zeer erkentelijk voor het beoordelen van mijn proefschrift.

De volgende mensen hebben voor mij uitstekend laboratoriumwerk verricht: Annemarie Boerma, bedankt voor de eindeloze cellijnexperimenten en medebeoordeling van de TMA's. Klaske ten Hoor, vanaf het allereerste begin was jij er om mij de fijne kneepjes van het microscopiewerk bij te brengen. Dank voor de vele immunohistochemische kleuringen die je hebt gedaan. En voor de gezellige gesprekken bij de koffie! Tineke van der Sluis, eindeloos heb je de IDO-antilichamen getest totdat we het juiste konden gebruiken. Daarnaast wil ik je bedanken voor jouw hulp bij het maken van de tissue microarray en je geduldige antwoorden op alle vragen die ik op je afvuurde.

Zonder de hulp van de volgende mensen was de totstandkoming van dit proefschrift een stuk ingewikkelder geweest: Gerard Kroon, dankzij jouw toegankelijkheid en expertise duurden problemen met computers gelukkig nooit lang. Bert Duitscher, bedankt voor de veelheid aan statussen die jij overal vandaan wist te halen. Janny Abels en Anna Slotboom, hartelijk dank voor jullie logistieke ondersteuning en voor de gezellige (koffie)pauzemomenten!

Mijn (ex)onderzoekscollega's van kamer Y4.234 en "de overkant" wil ik graag bedanken voor de gezellige koffie- en theepauzes, lunches, (wetenschappelijke) discussies en vrijdagmiddagborrels. Aniek, Annet, Catarina, Claudia, Elsbeth, Esther, Ingrid, Janna, Jasper, Justine, Maartje, Marjon, Marloes, Meike, Monique, Natalia, Ninke, Renee, Teelkien en Welmoed: dankzij jullie heb ik een hele gezellige promotietijd gehad.

De medewerkers van de afdeling radiotherapie van het UMCG wil ik bedanken voor hun begrip en flexibiliteit bij het afronden van mijn proefschrift. Vanaf nu zal ik met twee benen op de afdeling radiotherapie staan. En lieve collega AIOS, ik kijk uit naar de resterende opleidingstijd die ik met jullie mag doorbrengen. We gaan een stralende toekomst tegemoet!

Anne Hoekstra en Carla Niens, ik ben ontzettend blij dat jullie als paranimfen aan mijn zijde willen staan! Lieve Anne, vanaf de eerste dag van de geneeskunde opleiding is er een waardevolle vriendschap tussen ons ontstaan. Menigeen heeft gedacht dat wij zusjes zijn, wat zomaar zou kunnen gezien onze overeenkomsten. Jouw sarcastische humor en nuchtere levenshouding vind ik geweldig! Lieve Carla, wij weten elkaar altijd weer op te beuren met allerlei adviezen, om die vervolgens zelf niet op te volgen . . . Onze vriendschap vindt haar oorsprong in het mooie Friesland en werd versterkt tijdens onze studententijd. Ik ben blij dat we elkaar nog steeds weten te vinden.

Dankwoord

Lieve vrienden en (schoon)familie, door jullie warme belangstelling en afleiding gedurende stressvolle tijden heb ik mij enorm gesteund gevoeld. Vanaf nu zijn de weekenden weer volledig vrij voor borrels, etentjes, wandelingen en saunabezoekjes.

Lieve meiden van Roxie, zingen is voor mij volledige ontspanning. Jullie gezelligheid en vriendschap kreeg ik er gratis bij !

Leave heit en mem, in woorden valt niet uit te drukken hoeveel ik van jullie houd. Jullie hebben mij geleerd mijn eigen keuzes te maken, daarbij accepterend dat ik soms onderuit ging. Het besef dat ik altijd bij jullie terecht kan en volledig mezelf mag zijn is onbeschrijfelijk mooi.

Leave Rimmer, "broerke", wij beginnen we elkaar steeds meer te vinden. Ik ben trots erop hoe je je eigen (studie)keuzes hebt gemaakt en hebt doorgezet in lastige tijden. Ik heb het volste vertrouwen erin, dat jij een mooie toekomst tegemoet gaat.

Lieve Paul, bedankt voor jouw morele en taalkundige ondersteuning bij het schrijven van dit proefschrift. Dankzij jou begin ik de kunst van het relativeren en ontspannen langzamerhand (een beetje) onder de knie te krijgen. Jij bent de mooiste aanvulling op mijn leven. Ik hou van je!

Renske

167

- I

Publication overview


170


De Jong RA, Pras E, Boezen HM, van der Zee AGJ, Maurits MJE, Arts HJG, Aalders JG, Slot A, Timmer PR, Hollema H, Nijman HW. Less gastrointestinal toxicity after adjuvant radiotherapy on a small pelvic field compared to a standard pelvic field in patients with endometrial carcinoma. Int J Gynecol Cancer 2012;22:1177-86.

De Jong RA, Kema IP, Boerma A, Boezen HM, van der Want JJL, Gooden MJM, Hollema H, Nijman HW. Prognostic role of indoleamine 2,3-dioxygenase in endometrial carcinoma. Gynecol Oncol 2012;126:474-80.

De Jong RA, Nijman HW. Response to Letter to the Editor regarding the article entitled "status of cellular immunity lacks prognostic significance in vulvar squamous carcinoma", by R.A. de Jong et al. Gynecol Oneal. (2012), doi:10.1016/j .ygyno.2011 .12.416 for Gynecologic Oncology. Gynecol Oncol 2012;125:768-9.

De Jong RA, Boerma A, Boezen HM, Maurits MJE, Hollema H, Nijman HW. Loss of HLA class I and mismatch repair protein expression in sporadic endometrioid endometrial carcinomas. Int J Cancer 2012;131:1828-36.

De Jong RA, Toppen NL, ten Hoor KA, Boezen HM, Kema IP, Hollema H, Nijman HW. Status of cellular immunity lacks prognostic significance in vulvar squamous carcinoma. Gynecol Oneal 2012;125:186-93.

De Jong RA, Nijman HW, Boezen HM, Volmer M, ten Hoor KA, Krijnen J, van der Zee AGJ, Hollema H, Kema IP. Serum tryptophan and kynurenine concentrations as parameters for indoleamine 2,3-dioxygenase activity in patients with endometrial, ovarian and vulvar cancer. Int J Gynecol Cancer 2011;21:1320-7.

De Jong RA, Nijman HW, Wijbrandi TF, Reyners AKL, Boezen HM, Hollema H. Molecular markers and clinical behavior of uterine carcinosarcomas; focus on the epithelial tumor component. Mod Pathol 2011;24:1368-79.

Bijen CBM, Bantema-Joppe EJ, de Jong RA, Leffers N, Maurits MJE, Eggink HF, van der Zee AGJ, Hollema H, de Bock GH, Nijman HW. Classical and non classical MHC class I expression as prognostic factor in endometrial cancer. I nt J Cancer 2010;126:1417-27.

Jongen VHM, Briet JM, de Jong RA, Joppe E, ten Hoor KA, Boezen HM, Evans DB, Hollema H, van der Zee AGJ, Nijman HW. Aromatase, Cyclooxygenase 2, HER-2/neu, and p53 as prognostic factors in endometrioid endometrial cancer. Int J Gynecol Cancer 2009;19:670-6.

De Jong RA, Leffers N, Boezen HM, ten Hoor KA, van der Zee AGJ, Hollema H, Nijman HW. Presence of tumor-infiltrating lymphocytes is an independent prognostic factor in Type I and II endometrial cancer. Gynecol Oncol 2009;114:105-10.


Jongen V, Briet J, de Jong RA, ten Hoor K, Boezen M, van der Zee A, Nijman H, Hollema H. Expression of estrogen receptor-alpha and -beta and progesterone receptor-A and -B in a large cohort of patients with endometrioid endometrial cancer. Gynecol Oncol 2009;112:537-42.

Leffers N, Gooden MJM, de Jong RA, Hoogeboom BN, ten Hoor KA, Hollema H, Boezen HM, van der Zee AGJ, Daemen T, Nijman HW. Prognostic significance of tumor-infiltrating T-lymphocytes in primary and metastatic lesions of advanced stage ovarian cancer. Cancer Immunol Immunother 2009 ;58:449-59.

171

-

Curriculum vitae

Curriculum vitae

Curriculum vitae

Renske de Jong werd op 5 juli 1984 geboren in Heerenveen. In 2002 behaalde zij haar gymnasiumdiploma aan het Bornego College in Heerenveen. Hierna startle zij met de studie geneeskunde aan de Rijksuniversiteit Groningen waarvan ze de propedeuse haalde in 2003. In het vierde studiejaar deed zij haar wetenschappelijke stage op de afdeling Gynaecologische Oncologie van het Universitair Medisch Centrum Groningen met als onderwerp "Evaluation of estrogen and progesterone receptors as prognostic markers in endometrial cancer". In periode tussen het einde van de wetenschappelijke stage en de start van haar co-schappen reisde ze twee maanden door Zuid-Afrika, waar ze vrijwilligerswerk deed. De overige 4 maanden besteedde zij aan wetenschappelijk onderzoek. Dit vormde later de basis van haar promotieonderzoek. Renske heeft haar co-schappen doorlopen in de Isala Klinieken in Zwolle en Bethesda ziekenhuis in Hoogeveen. Gedurende deze periode schreef zij haar eerste artikel: "Presence of tumorinfiltrating lymphocytes is an independent prognostic factor in type I and II endometrial cancer" waarvan zij de resultaten mocht presenteren op het Internationale ESGO congres in Berlijn. Na het behalen van het artsexamen in 2009 startte zij op de afdeling Gynaecologische Oncologie van het Universitair Medisch Centrum Groningen officieel met haar promotieonderzoek.

Renske is op 1 juni 2011 begonnen met de opleiding tot radiotherapeut-oncoloog in het Universitair Medisch Centrum Groningen (opleider: dr. M.A.A.M. Heesters).

175

1),, 01 ()�5,j

Documents

· Stellingen behorende bij het proefschrift Molecular and biological nature of endometrial 1. Om de prognose van patienten met een uterien carcinosarcoom te verbeteren, toekomstige