Adjuvant Treatment for High-Risk Operable Prostate Cancer B

Home , Estramustine phosphate, Retroprogesterone

Journal of Cancer Therapy, 2010, 1, 1-47 Published Online March 2010 in SciRes (http://www.SciRP.org/journal/jct/).

TABLE OF CONTENTS

Volume 1 Number 1 March 2010

Synthesis, Characterization and Anti-Angiogenic Effects of Novel 5-Amino Pyrazole Derivatives on Ehrlich Ascites Tumor [EAT] Cells in-Vivo H. Raju, S. Chandrappa, M. K. Ramakrishna, T. S. Nagamani, H. Ananda, S. M. Byregowda, K. S. Rangappa……………………1 Adjuvant Treatment for High-Risk Operable Prostate Cancer B. Paule, N. Brion…………………………………………………………………………………………………………………10 Novel Ethyl 2-(1-aminocyclobutyl)-5-(benzoyloxy)-6-hydroxy-pyrimidine-4-carboxylate Derivatives: Synthesis and Anticancer Activities D. Asha, C. V. Kavitha, S. Chandrappa, D. S. Prasanna, K. Vinaya, S. C. Raghavan, K. S. Rangappa……………………………21 Transfusion of Ipscs-Derived Leukocytes to Kill Cancer J. Li, Y. Cui, G. D. Gao, T. F. Yuan………………………………………………………………………………………………29 Ovarian Sex Cord-Stromal Tumors in Postmenopausal Women and Total Laparoscopical Management A. Tinelli, M. Pellegrino, V. E. Chiuri, A. Malvasi………………………………………………………………………………31 Sum-Based Meta-Analytical Enrichment of Gene Expression Data to Identify Pathway Signatures of Cancers K. Wagholikar, P. Venkatraman, S. Vijayraghavan, C. Kumar-Sinha……………………………………………………………36 Role of Estradiol, Progestins, Insulines and Adipocytokines in Breast Cancer Promotion in Post-Menopausal Women C. Jamin…………………………………………………………………………………………………………………………43

SUBSCRIPTIONS

The Journal of Cancer Therapy (Online at Scientific Research Publishing, www.SciRP.org) is published quarterly by Scientific Research Publishing, Inc., USA.

Subscription rates: Print: $50 per issue. To subscribe, please contact Journals Subscriptions Department, E-mail: [email protected]

SERVICES

Advertisements Advertisement Sales Department, E-mail: [email protected]

Reprints (minimum quantity 100 copies) Reprints Co-ordinator, Scientific Research Publishing, Inc., USA. E-mail: [email protected]

All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as described below, without the permission in writing of the Publisher.

Copying of articles is not permitted except for personal and internal use, to the extent permitted by national copyright law, or under the terms of a license issued by the national Reproduction Rights Organization.

Requests for permission for other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works or for resale, and other enquiries should be addressed to the Publisher.

Statements and opinions expressed in the articles and communications are those of the individual contributors and not the statements and opinion of Scientific Research Publishing, Inc. We assumes no responsibility or liability for any damage or injury to persons or property arising out of the use of any materials, instructions, methods or ideas contained herein. We expressly disclaim any implied warranties of merchantability or fitness for a particular purpose. If expert assistance is required, the services of a competent professional person should be sought.

PRODUCTION INFORMATION

For manuscripts that have been accepted for publication, please contact: E-mail: [email protected] Journal of Cancer Therapy, 2010, 1: 1-9 1 doi:10.4236/jct.2010.11001 Published Online March 2010 (http://www.scirp.org/journal/jct)

Synthesis, Characterization and Anti-Angiogenic Effects of Novel 5-Amino Pyrazole Derivatives on Ehrlich Ascites Tumor [EAT] Cells in-Vivo

H. Raju1, S. Chandrappa1, M. K. Ramakrishna1, T. S. Nagamani1, H. Ananda2, S. M. Byregowda2, K. S. Rangappa1*

1Department of Studies in Chemistry, University of Mysore, Manasagangotri, India; 2Institute of Animal Health and Veterinary Biologicals, Hebbal, India. Email: [email protected], [email protected]

Received November 11th, 2009; revised December 14th, 2009; accepted December 25th, 2009.

ABSTRACT In search of synthetic chemotherapeutic substances capable of inhibiting, retarding, or reversing the process of multi-stage carcinogenesis, we synthesized a series of novel 5-amino pyrazole derivatives 11(a-h) by a nucleophilic substitution reaction and characterized by 1H nuclear magnetic resonance (NMR), liquid chromatography mass spec- trometry (LC/MS), Fourier-transform infrared (FTIR), and elemental analysis. These novel compounds were evaluated for their efficacy in inhibiting Ehrlich ascites tumor [EAT] cells in-vivo. In the present study we designed, synthesized, characterized and investigate the anti-angiogenic effects of these compounds, on Ehrlich ascites tumor [EAT] cells in-vivo. The compounds were subsequently tested for their ability to inhibit neovascularisation in chorio allantoin membrane (CAM) model. From the Structure Activity Relationship (SAR) studies, it reveals that, the substitution at N-terminal in pyrazole ring plays key role in the antitumor and anti-angiogenic effects.

Keywords: 1H-Pyrazol, Aryl Isothiocyanates, Ehrlich Ascites Tumor [EAT] Cells, Anti-Angiogenesis

1. Introduction undesirable side effects and development of drug resistance is the major clinical problem. Angiogenesis, a physiological process involving the Indian System of Medicine has many herbal prepara- growth of new blood vessels from pre-existing vessels, tions with versatile medicinal properties. In common contributes to the development and progression of vari- with many other low-molecular weight phenolic com- ous pathological conditions including tumor growth and pounds, vanillin displays antimicrobial and antioxidant metastasis, cardiovascular diseases, inflammatory disease properties and hence has the potential for use as a food and psoriasis. Down-regulation of angiogenesis has been preservative [2]. The chemo preventive effects of vanillin considered to be advantageous for prevention of neoplas- have been attributed to various biological properties in- tic growth and inflammation. Currently, anti-angiogenic cluding neutralization of carcinogenic free radicals [3] strategies are based on inhibition of endothelial cell pro- and anti-angiogenesis action [4,5]. There is some evi- liferation, interference with endothelial cell adhesion and dence for anti-mutagenic effects of vanillin, for example migration, and interference with metalloproteinases [1]. in suppressing chromosomal damage induced by meth- Many researchers have been trying to screen novel otrexate in the Chinese hamster V79 cell line [6]. Most anti-angiogenic principles from various natural products. studies have addressed the prognostic significance of Tumors can make and release many chemicals that can VEGF (Vascular endothelial growth factor) expression start angiogenesis. Using a drug that targets only one of [7]. Pyrazoles and several N-substituted pyrazoles are these chemicals may not have a large effect on the cancer, known to possess numerous chemical, biological, medici- but combining drugs that attack different targets may nal, and agricultural applications because of their versatile prove to be more useful. Synthetic compounds are used biological activities like antimicrobial activity [8], antitu- to control the advanced stages of malignancies but most mor and antileukemia activity [9], antidepressant and an- of the compounds exhibit normal tissue toxicity with ticonvulsant [10]. Amides are ubiquitous in life, as pro-

OH O Br O teins play a crucial role in virtually all biological processes O (i) O H + O + H2N N Boc such as enzymatic catalysis, transport or storage, immune O O O C O protection, and mechanical support. An in-depth analysis H O C H O of the comprehensive medicinal chemistry database re- 1 2 3 4 O O O vealed that the carboxamide group appears in more than O (ii) O O (iii) (iv) O 25% of known drugs. Recently we have reported the syn- O O O C H thesis and anti-angiogenisis studies of bioactive heterocy- H N N Boc H N N Boc H cles [11]. 5 6 O O O In the future, such drugs may blur the line between O O O O CH2CN O anti-angiogenesis drugs and other forms of cancer treat- O (v) + O O N ment. Researchers are now looking at many different H N N N NH2Cl aspects of anti-angiogenesis drugs. Better understanding H H N 9 2 7 8 of these drugs will probably make them a bigger part of F NO2 cancer treatment in the future. Hence there is a need to F R= (10a)= NO2 (10e)= O discover novel compounds that selectively kill cancer O cells. Recently, attention has been toward the drug de- O O (vi) (10b)= (10f)= F O rived from plant sources which are non-toxic and acces- 10(a-h) N N Cl Cl sible to common man. Hence attempts are made to syn- S (10c)= N (10g)= Cl thesize compounds of natural origin. R N H H Cl F 11 (a-h) 2. Materials and Methods (10d)= (10h)=

2.1 Chemicals and Reagents Figure 1. Reaction and reagent condition (a) K2CO3, DMF/ Unless otherwise mentioned all the chemicals used in the r.t, 20–30 min (b) EtOH /r.t, 2–3 h (iii) 10% Pd/c / H2 present study were from Sigma–Aldrich, USA. Melting EtOAc, r.t, 3h (c) dichloromethane, ether in HCl, 3 h (d) o points were determined using SELCO-650 hot stage melt- EtOH/EtCOONa, reflux 80 C, 2–3 h (e) triethylamine, di- ing apparatus and were uncorrected. Infrared (IR) spectra chloromethane, aryl isothiocyanates 10(a-h), r.t, 6–7 h were recorded using a Jasco FTIR-4100 series. Nuclear magnetic resonance (1H NMR and 13C NMR) spectra were was purified by column chromatography over silica gel recorded on Shimadzu AMX 400-Bruker, 400 MHz spec- (60–120 mesh) using chloroform: methanol (9:1) as an eluent. trometer using CDCl3 as solvent and TMS as internal standard (chemical shift in  ppm). Spin multiples are 2.1.2 Synthesis of 1H-Pyrazol-5-Amine (9) given as br s (broad singlet), d (doublet), t (triplet) and m Initially mono boc protected 3-methoxy-benzylidene (multiplet). Mass and purity were recorded on a LC/ hydrazine (5) was synthesised by the condensation MSD-Trap-XCT. Elemental (CHN) analyses were ob- reaction of 3-methoxy-benzaldehyde (3) (1.0 g, 9.84 tained on Vario EL III Elementar. Silica gel column chro- mmol) with mono boc protected hydrazine (4) (1.0 g, matography was performed using Merck 7734 silica gel 15.26 mmol). The subsequent double bond reduction was (60–120 mesh) and Merck made TLC plates. Substituted done by using 10% Pd/c in ethanol yielded mono boc protected 3-methoxybenzyl hydrazine (6). The deprotec- (E)-1H-pyrazol-5-amine 11(a-h) derivatives were synthe- tion of amine group was carried out by using HCl in sized by the method summarized in Figure 1. ether gave free amine compound (7). Finally the key in- 2.1.1 Synthesis of 3-Methoxybenzaldehyde (3) termediate 5-amino pyrazole (9) by the cyclisation of A solution of 4-hydroxy-3-methoxybenzaldehyde (1) 3-methoxybenzyl hydrazine salt (7) (1.0 g, 5.36 mmol) (2.0 g, 6.12 mmol) in N, N-dimethyl formamide (20 mL) and 3-cyclopropyl-3-oxopro-panenitrile (8) (0.85 g, 5.36 was taken, anhydrous potassium carbonate (4.2 g, 30.6 mmol) were taken in ethanol, and then sodium ethoxide mmol) was added and stirred for 20-30 min and then (1.09 g, 16.0 mmol) was added. The reaction mixture 5-(bromomethyl)-1, 2, 3-trimethoxybenzene (2) (2.69 g, was refluxed for 2–3 h. The progress of the reaction was 13.4 mmol) was added. The reaction mixture was stirred monitored by TLC. for 6 h at room temperature, and monitored by TLC. 2.1.3 General Procedure for the Synthesis and Upon completion, the solvent was removed under re- Characterization of 5-Amino Pyrazole duced pressure; residue was taken in water and extracted Derivatives 11(a-h) with ethyl acetate. Finally water wash was given to the organic layer and dried with anhydrous sodium sulphate. To the solution of intermediate compound 9 (1 eq) in The solvent was evaporated to get crude product which dichloromethane, triethylamine (3 eq) was added and

Copyright © 2010 SciRes JCT Synthesis, Characterization and Anti-Angiogenic Effects of Novel 5-Amino Pyrazole Derivatives 3 on Ehrlich Ascites Tumor [EAT] Cells in-Vivo cooled to 0–5 °C in an ice bath. Respective aryl isothio- with 10% ammonium chloride solution followed by wa- cyanate 10(a-h) (1 eq) were added at cold condition and ter wash and dried with anhydrous sodium sulphate. The stirred at room temperature for 6–7 h. The progress of the solvent was evaporated and the crude product obtained reaction was monitored by TLC. Upon completion of was purified by column chromatography over silica gel reaction water was added to reaction mixture and ex- (60–120 mesh) using hexane: ethyl acetate (8:2) as an tracted with ethyl acetate. The organic layer was washed eluent. Characterization of novel 5-amino pyrazole

Table 1. Elemental analysis, mp/(oC), physical state and yield of compounds 3, 9, 11(a–h)

Elemental analysis, Anal. Found (calcd)/% Compound Appearance m.p/(oC) Formula yield/% C H N

3 white solid 221-226 C 18H20O6 95 65.10(65.03) 6.04(6.02) - -

9 white solid 234-233 C 24H29N3O5 93 65.59(65.48) 6.64(6.62) 9.56(9.51)

11a white solid 254-256 C31H32N6O9S 89 56.02(56.01) 4.85(4.82) 9.64(9.61)

11b white solid 245-246 C31H34N4O5S 91 64.79(62.76) 5.96(5.93) 9.75(9.71)

11c white solid 263-265 C31H33ClN4O5S 88 61.10(61.08) 5.46(5.43) 9.18(9.11)

11d white solid 254-255 C31H33ClN4O5S 85 61.13(61.11) 5.48(5.45) 9.17(9.14)

11e white solid 248-249 C31H32F2N4O5S 81 60.97(60.08) 5.28(5.22) 9.17(9.11)

11f white solid 255-258 C31H33FN4O5S 84 62.82(62.80) 5.61(5.61) 9.45(9.41)

11g white solid 273-275 C31H32Cl2N4O5S 79 57.85(57.83) 5.01(5.0) 8.71(8.11)

11h white solid 267-268 C31H33FN4O5S 80 62.88(62.86) 5.67(5.63) 9.42(9.41)

Table 2. 1H NMR and IR data of compounds 3, 9, 11(a–h)

1 -1 Compound H NMR (CDCl3, 400 MHz) δ and IR (KBr, cm ) data 8.95 (s, 1H, -OH), 7.92 (dd, 1H, J = 2, 8.5 Hz, Ar-H), 7.18 (s, 1H, Ar-H), 6.82 (d, 1H, J = 8.5 Hz, Ar-H), 6.35 (br s, 1H, Ar-H), 3 6.15(br s, 1H, Ar-H), 5.71 (m, 2H, -CH2) 3.83 (s, 3H, -OCH3), 3.70 (s, 3H, -OCH3), 3.68 (s, 6H, -OCH3). 3328, 2912, 1665, 1583, 1502, 1443, 1266, 1124, 1017.

6.54 (s, 1H, Ar-H), 6.51 (dd, 1H, J = 2, 8.5 Hz, Ar-H), 6.46 (s, 1H, Ar-H), 6.15 (d, 2H, J = 6.5 Hz, Ar-H), 4.99 (m, 2H, -CH2), 9 3.95 (br s, 2H, -NH2), 1.85 (m, 1H, -CH), 0.92 (m, 2H, -CH2), 0.71 (m, 2H, -CH2). 3448, 2998, 2942, 2626, 1667, 1616, 1516, 1506, 1257, 1129, 1024, 1003, 924, 818, 772. 8.20–8.16 (dd, 1H, -NH), 8.06 (s, 1H, -NH), 7.65 (s, 1H, Ar-H), 7.11 (s, 1H, Ar-H), 7.52 (s, 1H, Ar-H), 6.1 (br s, 1H, Ar-H), 11a 5.21-5.18 (m, 2H, -CH2), 3.83 (s, 3H, -OCH3), 3.78 (s, 3H, -OCH3), 3.68 (s, 6H, -OCH3). 3448, 3318, 2998, 2902, 1665, 1584, 1506, 1455, 1414, 1276, 1123, 1027, 925, 851, 722. 8.20–8.16 (dd, 1H, -NH), 8.06 (s, 1H, -NH), 6.78 (s, 1H, Ar-H), 6.87 (s, 2H, Ar-H), 6.62 (s, 1H, Ar-H), 6.60 (s, 1H, Ar-H), 6.1 11b (br s, 1H, Ar-H), 5.21-5.18 (m, 2H, -CH2), 3.83 (s, 3H, -OCH3), 3.78 (s, 3H, -OCH3), 3.70 (s, 3H, -OCH3), 3.68 (s, 6H, -OCH3). 3001, 2929, 1665, 1598, 1583, 1506, 1272, 1127, 1028, 992, 862, 836, 778. 8.20–8.16 (dd, 1H, -NH), 8.06 (s, 1H, -NH), 7.52 (s, 1H, Ar-H), 6.89 (s, 1H, Ar-H), 6.56 (s, 1H, Ar-H), 6.40 (s, 1H, Ar-H), 6.1 11c (br s, 1H, Ar-H), 5.21-5.18 (m, 2H, -CH2), 3.83 (s, 3H, -OCH3), 3.78 (s, 3H, -OCH3), 3.68 (s, 6H, -OCH3). 3068, 2942, 1676, 1584, 1510, 1411, 1257, 1222, 1126, 1035, 913, 837, 714. 8.20–8.16 (dd, 1H, -NH), 8.06 (s, 1H, -NH), 6.95 (s, 1H, Ar-H), 6.63 (s, 1H, Ar-H), 6.47 (s, 1H, Ar-H), 6.34 (s, 1H, Ar-H), 6.1 11d (br s, 1H, Ar-H), 5.21-5.18 (m, 2H, -CH2), 3.83 (s, 3H, -OCH3), 3.78 (s, 3H, -OCH3), 3.68 (s, 6H, -OCH3). 3448, 2941, 1666, 1583, 1536, 1413, 1266, 1128, 1088, 1015, 996, 830, 730. 8.20–8.16 (dd, 1H, -NH), 8.06 (s, 1H, -NH), 6.72 (s, 1H, Ar-H), 6.21 (s, 1H, Ar-H), 6.18 (s, 1H, Ar-H), 6.1 (br s, 1H, Ar-H), 11e 5.21-5.18 (m, 2H, -CH2), 3.83 (s, 3H, -OCH3), 3.78 (s, 3H, -OCH3), 3.68 (s, 6H, -OCH3). 3448, 3318, 2998, 2935, 1769, 1675, 1667, 1588, 1462, 1411, 1278, 1249, 1159, 1124, 1023, 994. 8.20–8.16 (dd, 1H, -NH), 8.06 (s, 1H, -NH), 6.72 (s, 2H, Ar-H), 6.44 (s, 2H, Ar-H), 6.13 (s, 1H, Ar-H), 6.1 (br s, 1H, Ar-H), 11f 5.21-5.18 (m, 2H, -CH2), 3.83 (s, 3H, -OCH3), 3.78 (s, 3H, -OCH3), 3.70 (s, 3H, -OCH3), 3.68 (s, 6H, -OCH3). 3316, 2937, 1767, 1672, 1588, 1503, 1460, 1413, 1245, 1169, 1126, 1025, 991. 8.20–8.16 (dd, 1H, -NH), 8.06 (s, 1H, -NH), 6.96 (s, 1H, Ar-H), 6.41 (s, 1H, Ar-H), 6.28 (s, 1H, Ar-H), 6.1 (br s, 1H, Ar-H), 11g 5.21-5.18 (m, 2H, -CH2), 3.83 (s, 3H, -OCH3), 3.78 (s, 3H, -OCH3), 3.68 (s, 6H, -OCH3). 3448, 3207, 2998, 2942, 1667, 1616, 1516, 1506, 1278, 1257, 1129, 1024, 1003, 924, 818, 772. 8.20–8.16 (dd, 1H, -NH), 8.06 (s, 1H, -NH), 6.78 (s, 1H, Ar-H), 6.72 (s, 1H, Ar-H), 6.60 (s, 1H, Ar-H), 6.44 (s, 1H, Ar-H), 6.1 11h (br s, 1H, Ar-H), 5.21-5.18 (m, 2H, -CH2), 3.83 (s, 3H, -OCH3), 3.78 (s, 3H, -OCH3), 3.70 (s, 3H, -OCH3), 3.68 (s, 6H, -OCH3). 3318, 2935, 1769, 1675, 1588, 1506, 1462, 1411, 1249, 1159, 1124, 1023, 994.

derivatives are tabulated in Tables 1 and 2. The tumor inhibitory effects of the compounds on EAT cell growth were assessed by measuring cell number and 2.2 Biology: in-Vivo Anti-Cancer and ascites volume. On day 12 injected 0.5ml saline into Angio-Inhibitory Effects of Synthetic Novel abdomen. The control and compounds treated tumor 5-Amino Pyrazole Derivatives 11(a-h) bearing mice were sacrificed, an incision was made in Animals and tumor model: Inbred Swiss albino mice, 6–8 the abdominal region and EAT cells along with the ascites fluid were harvested into a beaker containing 2 ml saline weeks old, weighing 25 ± 5 g of either sex, were used for 0 the experiments. They were bred and maintained in the and centrifuged at 3000 rpm for 10 min at 4 C. Subtracting animal house, Institute of Animal Health and Veterinary the volume of saline added previously from the volume of the supernatant gave the volume of ascites fluid. After Biologicals, Hebbal, Bangalore, India. Ehrlich ascites tu- harvesting the EAT cells, the cells were resuspended in mor was grown in adult Swiss albino mice intraperitoneally 0.9% saline and counted using a haemocytometer. (ip). Cell viability was tested by trypan blue exclusion as- Changes in the morphology of EAT cells with Giemsa say. Experimental animals were prepared by injecting 5 × 6 staining: EAT cells from the control and treated groups 10 viable tumor cells into intraperitoneal cavity of Swiss 11(a–h) were smeared on clean glass slides, air-dried, mice. Tumor growth was followed by recording the animal and fixed in a solution of methanol/acetic acid (3:1). The weights. EAT cells begin their exponential growth phase th slides were hydrated with PBS, then stained with 0.1% from the 7 day after tumor cell injection and the animal Giemsa solution, and observed under compound micro- succumbs to the ascites tumor burden on day 16–20 after scope. injection. Acridine orange/ethidium bromide staining: Nuclear Compounds: Synthetic 5-amino pyrazole derivatives staining was performed according to the method of 11(a-h) were used as compounds for the experiments. The Srinivas [13]. The EAT cells collected from both control compounds were weighed and dissolved in 0.1% DMSO to and compounds 11(a–h) treated groups were smeared on get required concentrations. The compound treatments clean glass slides, air-dried, and fixed in a solution of were initiated on the day 7 of tumor transplantation on the methanol/acetic acid (3:1). The slides were hydrated with advanced stage of tumor when the cells enter into exponen- PBS, and then stained with mixture of acridine or- tial growth period. ange/ethidium bromide (1:1). The cells were immediately Animal survival: After 7 days of tumor cell injection, washed with PBS and viewed under fluorescent micro- the animals were divided into groups of 10 each and scope and photographed. were treated as follows: control: 0.2 ml of 0.1% DMSO Anti-angiogenic effects of the compounds in peritoneal was given on day 7, 9, and 11 of tumor transplantation. angiogenesis: The peritoneum of the mice was cut open Compound treated groups- the compounds 11(a-h) were and the inner lining of the peritoneal cavity were exam- given to different groups of tumor bearing mice. The ined for angiogenesis in both control and compounds compound 100 mg/kg body wt was injected intraperito- 11(a–h) treated tumor bearing mice and photographed. neally (ip) into the mice using 26 gauge needle on day 7, Histopathology of mice peritoneum tissue: Peritoneal 9, and 11 of tumor transplantation. All the mice were tissues from tumor bearing control mice and mice treated weighed on the day of tumor inoculation and at weekly with compounds 11(a–h) were fixed in 10% formalin, intervals. Animal survival was recorded up to 40 days. embedded in paraffin, and 5-μm sections were routinely The tumor response was assessed on the basis of MST stained with haematoxylin and eosin. The sections were and increase in life span (% ILS). Median survival time observed under low power (10X) of light microscope to (MST) and % ILS were calculated from the mortality identify the highly vascularized areas. The micro vessel data within the observation period. Increase in life span density (MVD) was counted in 10 fields of these vascu- was calculated by the formula. larized areas under high power (40X) and the average % ILS = MST of treated group - MST of control group×100 MVD/HPF was noted. MST of control group Angioinhibitory effects of the compounds on in vivo Enhancements of life span by 25% are more over that of chorio allantoin membrane assay: CAM was performed the control was considered as effective antitumor res- according to the method of Chandru and Sharada [14]. The ponse [12]. fertilized eggs were divided into different treatment groups. Tumor growth inhibition and anti-angiogenesis: After Control, the saline treated group, and compound treated 7 days of tumor cell injection, the animals were divided groups with minimum of six eggs in each group were into eight groups of 10 each and the control group maintained separately and observed. The fertilized eggs received 0.2 ml of 0.1% DMSO on day 7, 9, and 11 of were incubated for 5 days at 37 0C in a humidified and tumor transplantation. The compounds 11(a-h) were sterile atmosphere. A window was made under aseptic given to eight different groups of tumor bearing mice. conditions on the egg shell to check for the proper de-

Copyright © 2010 SciRes JCT Synthesis, Characterization and Anti-Angiogenic Effects of Novel 5-Amino Pyrazole Derivatives 5 on Ehrlich Ascites Tumor [EAT] Cells in-Vivo velopment of the embryo. The windows were resealed cantly higher in compound treated groups compared to and incubation was continued. On day 11 the windows control (Table 3) indicating the effect of the compounds were opened and the compounds 11(a–h) (0.1 mM) or in preventing the tumor growth. The treated groups vehicle was loaded on the cover slips separately, showed reduction in body weight due to decrease in tu- air-dried, and inverted over the CAM and the windows mor burden however no side effects were observed. were closed. The windows were resealed and the embryo was allowed to develop further. The windows were 3.2 Ascites Volume and Cell Number (Figure 2) opened and observed on day 13 and inspected for The inhibitory effect of 11(a-h) on EAT cells in vivo was changes of MVD in the area under cover slip and exam- further examined in terms of total volume of ascites and ined under a microscope for avascular zone and photo- number of cells in mice treated with vehicle or compounds. graphed. The mean value of cell number and ascites volume in con- Statistical analysis: All data were analyzed by trol animals was found to be 1820.30 ± 0.65×106 cells/ one-way ANOVA. Values of p < 0.05 were considered mouse (Figure 2(a)) and 9.24 ± 0.30 ml, respectively significant. (Figure 2(b)). All the compound treatments showed sig- 3. Results nificant decrease in ascites volume and cell number compared to control (p<0.01). The compound 11a treated 3.1 Tumor Growth Inhibition of Ehrlich Ascites group which demonstrated highest % ILS (Table 3) Tumor in-Vivo showed maximum reduction in cell number and ascites volume compared to other compounds treated groups. Animal survival (Table 3): The vehicle 0.1% DMSO did not have any effect on the tumor growth. All the animals 3.3 Changes in the Morphology of EAT Cells in vehicle treated controls developed tumor and died (Figure 3) within 16–20 days. The median survival time (MST) was 18 days. All the treatments produced significant increase in MST and % ILS compared to control (p < 0.01). Three doses of 11(a-h) treatments on 7, 9, and 11 days after tumor transplantation showed effective antitumor response (>25% ILS) and resulted in 48%, 34%, and 28% ILS, respectively. However the compound 11a exhibited higher tumor inhibitory effect and showed 32 days of MST with 77% ILS. The weight changes were signifi-

Table 3. Effect of synthetic novel 5-amino pyrazole derivatives 11(a-h) on survival of mice bearing Ehrlich ascites tumor (a) Treatment groups MST (days)a ILS (%) Av. wtb changes

1. Control 18 ------+ 11.32 (0.1% DMSO)

2. 11a-100 32.8** 77.7 - 4.4** 3. 11b-100 28.7** 68.2 - 3.2** 4. 11c-100 27.5** 70.6 - 2.8** 5. 11d-100 28.2** 71.9 - 2.9** 6. 11e-100 31.2** 77.2 - 4.2 ** 7. 11f-100 29.6** 73.2 - 3.1** 8. 11g-100 30.3** 76.8 - 3.6 ** (b) 9. 11h-100 23.9** 63.8 - 2.3** Figure 2. Effects of synthetic compounds 11(a-h) on ascites volume and cell number of EAT bearing mice. The bar 11(a–h)-100 indicates dose in mg/kg body wt. Vehicle or compounds were administered on day 7, 9, and 11 after tumor cell inoculation to graph represents the effect of the compounds on ascites different treatment groups of 10 animals each. aMedian survival time volume (a) and cell number (b). All the treatments showed a (MST) and ILS % was calculated from the mortality data within the significant difference in ascites volume and cell number observation period. bDetermined on 12th day of treatment.** Significant from the control and DMSO (p < 0.01). The error bars rep- from control (p < 0.01). resent standard deviation of the mean

(a) (a)

(b)

Figure 3. Changes in the morphology of EAT cells after (b) treatment with synthetic 5-amino pyrazole derivatives. EAT Figure 4. Suppression of angiogenesis in-vivo by compounds cells with or without compounds 11(a-h) treatment in-vivo 11(a-h). Peritoneal lining of tumor bearing mice treated were washed with PBS, fixed, and stained with Giemsa stain with vehicle (0.1% DMSO) and were inspected for (a) or with mixture of acridine orange: ethidium bromide (b). anti-angiogenesis effects. (a) Inhibitions of angiogenesis The cells were viewed under compound or fluorescent mi- were prominent in compound treated mice compared to croscope and photographed. The apoptotic bodies and con- control. (b) CAM assay model-compounds 11(a-h) or the densed nuclei are evident in the compound treated groups vehicle was applied on the CAM of 11 days old chick embryo. Decreased vasculature was observed in treated groups The inhibitory effect of synthetic 11(a-h) on EAT cell compared to control. Dotted circles indicate the area cov- growth may be due to induction of apoptosis. The EAT ered by the cover slip cells were stained with Geimsa or with nuclear stain (acridine orange: ethidium bromide) and the slides were observed under microscope and photographed. The apoptotic bodies and nuclear condensation are evident in compound treated groups. 3.4 Inhibition of Tumor Induced Neovascularisation (Figures 4(a) and 5)

Significant inhibitions of blood vessel formation were ob- (a) served in the peritoneal wall of compound treated mice compared to control (Figure 4(a)). The MVD (Figure 5) studies using haematoxylin and eosin stained peritoneal wall sections demonstrated significant decrease in MVD count in all the compounds 11(a–h) treated groups compared to control (p < 0.01). The strong angioinhibitory effect of the compounds was evident in this study. 3.5 Angioinhibitory Effects on CAM Assay (Figure 4(b)) The anti-angiogenic activity was evaluated by observing (b) the formation of avascular zone under the cover slip Figure 5. Histopathology and MVD count of the peritoneal (Figure 4(b)), control CAM treated with 0.1% DMSO wall sections of EAT bearing mice with or without in-vivo showed no changes in vasculature. The MVD decreased treatment of compounds 11(a-h). The arrow indicates the in the entire compounds 11 (a-h) treated CAM model. blood vessel present in section (a). Haematoxylin and eosin staining of peritoneal wall sections were observed to study 4. Discussion MVD. The bar graph shows the effect of compounds on MVD count (b) of the peritoneal section of tumor bearing The basic principle of cancer chemotherapy is to use mice. All the compound treatments resulted in a significant drugs that have targets and preferably non overlapping decrease in MVD count compared to control (p < 0.01). The toxicity. Thus a logical chemotherapeutic strategy is the error bars represent standard deviation of the mean

Copyright © 2010 SciRes JCT Synthesis, Characterization and Anti-Angiogenic Effects of Novel 5-Amino Pyrazole Derivatives 7 on Ehrlich Ascites Tumor [EAT] Cells in-Vivo combined use of apoptosis inducing compounds and cy- assay model compound 11a, 11e and 11g induced vasculature totoxic agents [15–17]. Examination of 11(a-h) aromatic zone formation in the developing embryos. Notably, a regions might be critical for potential ligand-receptor bind- newly formed microvessel was regressed around the ing. A reasonable approach following standard medicinal compounds implanted disc (Figure 4(b)). Quantification chemistry design concepts was to explore compounds of VEGF (Figure 6) shows that compounds 11(a-h) has with systematic differences in the carbon chain connect- dose dependent effect on secretion of VEGF under ing to the aromatic regions and substitution of cyclopro- in-vivo conditions compared to the untreated EAT bearing pyl ring and the nitro, chloro, fluoro phenyl groups in mice. The amount of VEGF increased in untreated EAT pyrazole ring led to the synthesis of eight novel com- cells over the growth period, whereas the amount of pounds. The results on Ehrlich ascites tumor [EAT] VEGF in ascites of compounds 11(a-h) treated EAT cells showed that the administration of 11(a-h) on day 7, 9, did not show any significant increase in the same growth and 11 after tumor cell inoculation produced effective period, suggesting a dose dependent inhibition of VEGF antitumor response (>25% ILS). The compounds also secretion upon compounds treatment in EAT cells. Fur- exhibited corresponding reduction in mean ascites volume ther, VEGF is considered to be one of the major stimula- and cell number. Our findings demonstrated the potent tors of tumor angiogenesis [21]. anti-angiogenic activity of the 5-amino pyrazole derivatives The compounds 11(a–h) with minor structural differ- against EAT in-vivo. The compounds effectively reduced ences has exhibited varying degree of tumor growth in- the ascites tumor burden and produced no side effects. hibition and anti-angiogenic activities against EAT Although the antitumor property of pyrazole and its in-vivo. In the present findings, the compound 11a with synthetic derivatives on in-vivo and in-vitro studies has the bioactive nitro, cyclopropoxy and methoxy groups at been reported earlier [18]. Morphology of EAT cells ortho, para and meta-positions in rings A and B showed after compound treatments exhibited apoptotic bodies, potent in-vivo antitumor and anti-angiogenic activities nuclear condensation, and intra nucleosomal fragmentation. against mouse tumor, Electron withdrawing groups such These results indicate that the compounds 11(a–h) as dinitro, difluoro and dichloro at para and meta position induced inhibition of EAT cell growth was due to the in 11a, 11e and 11g showed relatively significant activity, induction of apoptosis in EAT cells. As the compounds whereas 11(c, d, f, h) groups having fewer electrons are injected (ip) directly into the peritoneal cavity where withdrawing groups of chloro (ortho), (meta) and fluoro the tumor is growing, the effect would be immediate and (para), (ortho) showed moderate activity. On the other direct on the tumor cells. In the control EAT bearing hand, as the electron withdrawing efficiency increases, mice; extensive peritoneal angiogenesis was observed the activity also increases. Compound 11b without sub- which may be due to the secretion of the angiogenesis stituent on the phenyl ring of aryl isothiocyanates inducing factors in the ascites fluid. Involvement of showed poor activity. We have briefly investigated the VEGF in the formation of malignant ascites has been different structure-activity relationships (SAR) of the well documented [19]. Treatment of synthetic 11(a–h) to aryl isothio cyanates 11(a-h) functionalized derivatives EAT bearing mice significantly decreased peritoneal angio- with different groups added on the phenyl ring. These genesis suggesting the inhibition of the secretion of such modifications change the potency of anti-angiogenic ac- factors and thereby preventing the formation of new tivity profile of the synthesised compounds. Thus the blood vessels. Micro vessel density (MVD) counts have structural modifications have profound influence on an- become the morphological gold standard to assess the titumor and angioinhibitory activities of compounds. neovasculature in tumors. MVD counts are reflective of From our findings we can conclude that 11a > 11e > 11g the angioarchitectural properties of the tumor in that they are a representative of the average intercapillary distance. This is in fact an important parameter as it is the goal of an anti-angiogenic tumor therapy to reduce the intercapillary distance to a degree that it becomes rate-limiting for the growth of the tumor [20]. Haematoxylin and eosin staining of peritoneal wall tissue sections of EAT bearing mice treated or untreated with the compounds 11(a–h) were examined for the MVD count. Angiogenesis is clearly evident in the inner peritoneal lining of EAT bearing mice and it is a reliable model for in-vivo angiogenesis. Hence the peritoneal linings of treated mice verified Figure 6. Effect of compounds on in-vivo production of for its effect on microvasculature when compared to VEGF. The bar graph shows the amount of VEGF in ascites untreated EAT bearing mice (Figure 4(a)). In the CAM of compounds 11(a-h) treated EAT cells

Copyright © 2010 SciRes JCT 8 Synthesis, Characterization and Anti-Angiogenic Effects of Novel 5-Amino Pyrazole Derivatives on Ehrlich Ascites Tumor [EAT] Cells in-Vivo novel 11(a-h) can be considered as promising antitumor [5] C. Keshava, N. Keshava, O. Tong-man, and N. Joginder, and anti-angiogenic compounds. Thus our results show “Protective effect of vanillin on radiation-induced micro- that compounds having electron withdrawing groups nuclei and chromosomal aberrations in V79 cells,” Mutation Research/Fundamental and Molecular Mecha- have more potent activity. Hence, we compare the results nisms of Mutagenesis, Vol. 397, pp. 149–159, 1998. within same electron withdrawing nitro, fluoro and chloro group present at different position. Currently, a [6] T. S. Daniel, S. Woodrow, and M. D. David, “The an- timutagenic effect of vanillin and cinnamaldehyde on large variety of chemotherapeutic drugs are being used to spontaneous mutation in Salmonella TA104 is due to a treat cancer. Unfortunately, many compounds hold lim- reduction in mutations at GC but not AT sites,” Mutation ited efficacy, due to problems of delivery and penetration, Research/Fundamental and Molecular Mechanisms of and a moderate degree of selectivity for the tumor cells, Mutagenesis, Vol. 480, pp. 55–69, 2001. thereby causing severe damage to healthy tissues. From [7] M. Baba, H. Konno, and S. B. Nakamura, “Relationship our studies, it is clear that compound 11a, 11e and 11g of p53 and vascular endothelial growth factor expression has antiangiogenic effect as shown by peritoneal angio- to clinicopathological factors in human scirrhous gastric genesis assay, chorioallantoic membrane (CAM) assay, cancer,” European Surgical Research, Vol. 30, pp. 130– and also from the reduction in the EAT cell number, as- 137, 1998 cites volume, and body weight of the animals in-vivo. [8] E. Akbas and I. Berber, “Antibacterial and antifungal The above study shed light toward the identification of activities of new pyrazolo [3, 4-d] pyridazin derivatives,” new anti-angiogenic molecules to the cancer therapy. European Journal of Medicinal Chemistry, Vol. 40, pp. 401–408, 2005. Further research to know the mechanism of inhibition and the modifications of the compounds 11a, 11e and [9] S. Y. Fulmer and O. C. Allen, “Synthesis, antileukemic activity, and stability of 3-(substituted-triazeno) pyrazole 11g to improve the potency is currently under progress in -4-carboxylic acid esters and 3-(substituted-triazeno) pyr- our laboratory. azole-4-carboxamides,” Journal of Pharmaceutical Sci- 5. Acknowledgement ence , Vol. 60, No. 4, pp. 554–560. 1970. [10] Z. Ozdemir, H. B. Kandilici, B. Gumusel, U. Calis, and A. One of the authors H.Raju is grateful to UGC Research Bilgin, “Synthesis and studies on antidepressant and an- Fellowships in Sciences for Meritorious Students Scheme ticonvulsant activities of some 3-(2-furyl)-pyrazoline de- (RFSMS), New Delhi for financial support under rivatives,” European Journal of Medicinal Chemistry, Vol. RFSMS-JRF order No. DV5/373[13]-/RFSMS/ 2008–09. 42, pp. 373–379, 2007 The 1H and 13C, CHN, IR and other data obtained from [11] B. S. Priya, C. Anil Kumar, S. NanjundaSwamy, Bas- the instruments purchased under DST-FIST and appa., S. Naveen, J. ShashidharaPrasad, and K. S. Ran- UGC-SAP (phase II) Programmes are greatly acknowl- gappa, “2-(2-(2-Ethoxybenzoyl amino)-4-chlorophenoxy)- edged. N-(2-ethoxybenzoyl) benzamine inhibits EAT cell induced angiogenesis by down regulation of VEGF secretion,” Bioorganic & Medicinal Chemistry Letters, Vol. 17, pp. REFERENCES 2775–2780, 2007. [1] A. W. Griffioen and G. Molema, “Angiogenesis: Poten- [12] A. C. Sharada, F. E. Solomon, P. U. Devi, N. Udupa, and tials for pharmacologic intervention in the treatment of K. Srinivasan, “Antitumor and radiosensitizing effects of cancer, cardiovascular diseases, and chronic inflamma- withaferin A on mouse ehrlich ascites carcinoma in vivo,” tion,” Pharmacol Reviews, Vol. 52, No. 2, pp. 237–268, Acta Oncologica, Vol. 35, pp. 95–102, 1996. 2000. [13] G. Srinivas, R. JohnAnto, P. Srinivas, S. Vidyalakshmi, V. [2] T. Vaios, Karathanos, and I. Mourtzinos, “Study of the Priya Senan, and D. Karunagaran, “Emodin induces solubility, antioxidant activity and structure of inclusion apoptosis of human cervical cancer cells through poly complex of vanillin with β-cyclodextrin,” Food Chemis- (ADP-ribose) polymerase cleavage and activation of cas- try, Vol. 101, pp. 652–658, 2007. pase-9,” European Journal Pharmacol, Vol. 473, pp. 117– 125, 2003. [3] A. Soumyakanti, I. P. Kavirayani, and T. Mukherjee, “Physico-chemical studies on the evaluation of the anti- [14] C. AnilKumar, Basappa, P. S. Bharathi, and K. S. Ran- oxidant activity of herbal extracts and active principles of gappa, “Pro-apoptotic activity of imidazole derivatives some indian medicinal plants,” Journal of Clinical Bio- mediated by up-regulation of Bax and activation of CAD chemistry Nutrition, Vol. 40, No. 3, pp. 174–183, 2007. in Ehrlich Ascites Tumor cells,” Invest New Drugs, Vol. 25, pp. 343–350. 2007. [4] L. Kriengsak, K. Jan-Peter, and S. Tuangporn, “Vanillin suppresses metastatic potential of human cancer cells [15] Z. Franco, P. Paola, P. Silvana, P. Graziella, S. Rosanna, through PI3K inhibition and decreases angiogenesis in and A. Federico, “Role of apoptotic response in cellular vivo,” Journal of Agricultural and Food Chemistry, Vol. resistance to cytotoxic agents,” Pharmacology & Thera- 57, No. 8, pp. 3055–3063, 2009. peutics, Vol. 76, No. 1, pp. 177–185, 1997.

[16] K. Tetsuro, S. Kazunari, K. Hideaki, and M. Yukihisa, [19] T. Browder, C. E. Butterfield, B. M. Kraling, B. Shi, B. “Enhanced suppression of tumor growth by combination Marshall, S. Michael, M. S. O’Reilly, and Folkman, of angiogenesis inhibitor O-(Chloroacetyl-carbamoyl) “Antiangiogenic scheduling of chemotherapy improves fumagillol (TNP-470) and cytotoxic agents in mice,” efficacy against experimental drug-resistant cancer,” Cancer Research, Vol. 54, pp. 5143–5147, 1994. Journal of Cancer Research, Vol. 60, pp. 1878–1886, [17] B. A. Teicher, S. A. Holden, G. Ara, T. Korbut, and K. 2000. Manon, “Comparison of several antiangiogenic regimens [20] J. W. Rak, B. D. St Croix, and R. S. Kerbel, “Conse- alone and with cytotoxic therapies in Lewis lung carci- quences of angiogenesis for tumor progression, metastasis noma,” Cancer Chemothe Pharmacol, Vol. 38, pp. 169– and cancer therapy,” Anticancer Drugs, Vol. 6, No. 1, pp. 177, 1996. 3–18, 1995. [18] S. Chandrappa, H. Chandru, A. C. Sharada, K. Vinaya, C. [21] Z. J. Shang, Z. B. Li, and J. R. Li, “VEGF is up-regulated S. AnandaKumar, N. R. Thimmegowda, P. Nagegowda, by hypoxic stimulation and related to tumour angiogene- M. Karuna Kumar, and K. S. Rangappa, “Synthesis and sis and severity of disease in oral squamous cell carci- in vivo anticancer and antiangiogenic effects of novel noma: In vitro and in vivo studies,” International Journal thioxothiazolidin-4-one derivatives against transplantable of Oral and Maxillofacial Surgery, Vol. 35, pp. 533–538, mouse tumor,” Medicinal Chemistry Research (In press), 2006 2009.

Adjuvant Treatment for High-Risk Operable Prostate Cancer

Bernard Paule1, Nathalie Brion2,3

1AP-HP, Groupe Henri Mondor-Albert Chenevier, Service Oncologie, Créteil, France; 2Centre Hospitalier de Versailles, Hôpital André Mignot, unité Thérapeutique, Le Chesnay, France; 3UFR de Médecine Paris Ile-de-France Ouest, Université Versailles Saint Quentin en Yvelines, Guyancourt, France. Email: [email protected]

Received December 14th, 2009; revised December 25th, 2009; accepted December 28th, 2009.

ABSTRACT Patients who have undergone a radical prostatectomy may have to face high risks of recurrence. The risk of recurrence is elevated due to probable occult metastatic disease at the time of diagnosis. A rationale for using multimodal approach in order to minimize the chance of disease recurrence and to improve the survival of high risk patients is emerging from preclinical and clinical studies. New molecular and genetics assays, may help to select patients most likely to benefit from these approaches. In this review, we will especially discuss the potential benefits of adjuvant therapy after radical prostatectomy. This paper presents the identification of these high-risk patients; the explanation of an adjuvant treatment of residual disease after a radical prostatectomy; the clinical studies with adjuvant androgen deprivation, radiotherapy and/or chemotherapy and the microarrays analysis. This review highlights the importance of these new adjuvant treatments that aims at targeting the factor which triggers metastatic disease following a radical prostatectomy.

Keywords: Adjuvant Treatment, Radical Prostatectomy, High-Risk Patients

1. Introduction (ADT), chemotherapy, prostate bed radiotherapy (RT) or some combination of these modalities. According to the pre-operative d’Amico criteria, patients with localized prostate cancer (Pca) (PSA>20 ng/mL, 2. Identifying High-Risk Patients Gleason 8-10, T2c to T4 disease) are considered to be at According to CaPSURE study and using D’Amico’s cri- high risk, with recurrence rates ranged from 50 to 100 teria, around 20 to 30 percent of localized prostate can- percent after a local therapy alone, especially if they are cers would be at high-risk of progression [1,7] and, as young, healthy and with a long life expectancy. For these well, about 30% to 35% of non metastatic prostate can- patients, prostate cancer specific survival is significantly cers will eventually relapse with distant disease [8]. compromised [1] and surgery alone won’t be able to con- High-risk Pca has higher biochemical relapse or disease trol the disease. Instead, these patients can show signs of recurrence rate after RP. Prior surgery, the identification residual disease at the primary site with likely persistent of such aggressive cancers can be based on, at least, three androgen-dependent and independent subpopulation of well-defined predictors of the disease extent and outcome malignant cells. They also have high risk to develop as- after treatment: patients with clinical stage T3 or T4 dis- ymptomatic or symptomatic metastases. In this case, ad- ease, a serum PSA level of above 20 ng/ml, and those juvant approach may be especially important. It is well with Gleason scores of 8-10 plus some 4+3 Gleason known that, in breast or colon cancers, the use of adju- score but with negative bone scan and negative computed vant treatment after surgery has shown a beneficial im- tomography (CT) scan of abdomen and pelvis. In addi- provement in survival [2–6]. In Pca, randomized studies tion, a number of additional clinical parameters could are needed to evaluate the potential effect of adjuvant potentially be used to identify patients with high risk of therapy in these high-risk patients. The optimum adjuvant recurrence. Those includes PSA velocity of >2.0ng/mL/ management for high-risk patients after radical prostate year, at least 50% positive biopsies cores or either tissue ctomy (RP) may consist in androgen deprivation therapy cores invaded by tumor above 20% [9,10].

D’Amico reported in 2004 a study including 1.095 pa- gen-dependent cells. tients who underwent RP and who did not receive adju- 2) Among patients treated by RP with occult distant vant therapy [11]. At a median follow-up of 5.1 years, 27 diseases including metastases and micrometastases, an of 84 deaths were attributable to prostate cancer. On mul- early adjuvant hormone therapy may destroy the andro- tivariable analysis, preoperative PSA velocity>2ng/ gen-dependent residual tumour cells. By contrast, if the ml/year was associated with an increased risk of cancer number of residual tumour cells is too important, the specific mortality (RR: 0.8, p<0.01). This was also an presence of many androgen-independent clones could increased risk of overall mortality (RR: 0.9, p=0.01). make the hormone therapy ineffective and chemotherapy Pretreatment Gleason score 8 to 10 correlated with an necessary. Pound et al. [8] observed that patients relaps- increase in cancer-specific mortality (RR: 32, p=0.02). ing less than two years after RP had particular clinical Patients with higher clinical stage were at greater risk for and pathological characteristics: preoperative PSA>10 death from Pca (RR: 7.4, p=<0.01) and 2.0 (p=0.004) for ng/ml, Gleason>7 or pT3. Survival without progression death from any cause. was decreased and could justify an adjuvant treatment. Today, clinicians may be able to better characterize 3) Using Dunning R3327 rat prostatic adenocarcinoma high risk patients and predict the probability of Pca re- model that creates lung metastasis on untreated recipient currence for each patient through the use of several re- hosts, studies demonstrated that there was a direct rela- cently developed statistical models called nomograms [12, tionship between primary tumor size at the time of surgi- 13]. The presence of micrometastases remains a major cal removal and the number of lung metastases [23] This issue since it is likely that many high-risk Pca have mi- concept is in favor of early treatment after local therapy crometastasized at the time of diagnosis [14–16]. Though such as RP. Theoretically, when the tumor burden of an- the literature in this regard is poorly documented in pros- drogen independent cells is low, chemotherapy could be tate cancer, upcoming methods to detect those types of more effective. In other words, if treatment is delayed, microscopic diseases would help to decide appropriate the ability of adjuvant chemotherapy to cure the disease therapeutic strategies [17,18]. Finally, gene expression may be lost. These results emphasize the critical re- profiling of prostate carcinoma could be an alternative quirement of combining surgery and adjuvant chemo- means to distinguish aggressive tumor. Biology and inte- therapy as early as possible in the treatment of occult gration of gene expression signature together with clini- metastases, in order to minimize the total metastatic tu- cal variables may improve the outcome prediction for mor burden and maximize the possibility of cure. In hu- patients treated with RP [19]. man, in recent decades, several cytotoxic agents have been tested as monotherapy in metastatic hormone re- 3. Adjuvant Treatment fractory Pca with a certain success, at least in terms of PSA response and quality of life [24–33]. Even if these Adjuvant treatment is defined as an additional therapy drugs are still deficient as to cure hormone refractory given in association with primary management. RP alone disease, the observed effects strongly support their sig- cannot be considered as an efficient curative treatment nificant activity on distant disease. for locally advanced Pca, due to the high risk of regional 4) A study using the serially transplantable Dunning or distant metastases and local failure [20,21]. In these R-3327H rat prostatic adenocarcinoma has shown how conditions adjuvant treatment may be important so as to changing the timing of androgen ablation alone and of control the local and/or distant disease. Importantly sys- hormone-chemotherapy can affect the tumor growth rate temic adjuvant therapy will not compensate for insuffi- and host survival [34]. This study demonstrated three cient local therapy. basic points: a) when either androgen ablation or cytoxan 3.1 Rationale for Adjuvant Treatment chemotherapy were given as a single agent treatment, they were both more effective when given as early as Clinical data and preclinical models provide a rationale possible; b) when androgen ablation was combined with for adjuvant therapy and notably for the concomitant ad- cytoxan chemotherapy, it was more effective when both ministration of hormonal treatment and chemotherapy in therapies were begun simultaneously and as early as pos- prostate cancer. sible; and c) when androgen ablation and cytoxan treat- 1) In human prostate cancer xenografts, Craft et al. [22] ments were initiated simultaneously and early, it was have shown that prostate cancers contain heterogeneous possible to increase survival as compared with the groups mixture of cells that vary in their dependence on andro- who were given one of the two therapies alone (i.e., such gens for growth and survival, and that treatment with simultaneous early treatment enhanced the individual anti-androgen therapy provides a selective pressure. The therapeutic effectiveness of both treatments). latter stage of androgen independence could result from 5) Preclinical data evaluating the optimal timing and the clonal expansion of androgen-independent cells that combination of androgen deprivating therapy (ADT) in are present at a frequency of about 1 per 105-106 andro- LNCaP and Shionogi prostate cancer xenografts reported

Copyright © 2010 SciRes JCT 12 Adjuvant Treatment for High-Risk Operable Prostate Cancer that the mice that received simultaneous hormone-chemo- have reported on the beneficial outcome of adjuvant RT therapy had a significant improvement in time to treat- in patients with pathological risk factors after RP. The ment failure compared to sequential therapy. A marked EORTC 22911 trial showed that adjuvant RT (60 Gy) lack of response to castration was observed after initial was associated with improvement in biochemical pro- paclitaxel therapy. Moreover, transcriptional profiling gression free survival (74% versus 52.6% ; p<0.0001) [40] identified, after paclitaxel exposure, an increased expres- but the impact on overall survival awaits maturation on sion of several survival genes known to play a role in the data. In the South Western Oncology Group (SWOG) androgen independence [35]. These findings supported trial 8794 [41] which randomized 425 high-risk patients simultaneous chemohormonal therapy-adjuvant trials. to adjuvant RT after RP versus RP alone, no benefit in 3.2 Adjuvant Post RP Treatment terms of overall survival was observed in patients assigned to the adjuvant group. It was shown that adjuvant 3.2.1 Adjuvant Hormone Therapy (HT) radiation reduced the risk of biochemical treatment fail- Messing et al. have [36] demonstrated that adjuvant HT ure by 50% over RP alone. High-risk was defined as ex- significantly improves survival in patients with positive tracapsular tumor extension, positive surgical margins, or lymph nodes. Data were updated [37] regarding the use seminal vesicle involvement. To be eligible, patients had of immediate versus deferred ADT in patients found to to have histologically negative lymph nodes and a nega- have node-positive disease at the time of PR. At a median tive bone scan. Adjuvant radiation to the prostate bed (60 follow-up of 11.9 years, the survival results remain un- to 64 Gy) also seemed to reduce the risk of metastatic changed. The median survival for the immediate and dif- disease and biochemical failure at all postsurgical PSA fered ADT arms, was 13.9 years (2.1-14.5) and 11.3 levels [42]. Of note, in this study, the pattern of treatment years (1.3-14.2) respectively. The median disease specific failure in high-risk patients was predominantly local with survival has not been reached in the immediate arm yet a surprisingly low incidence of metastatic failure. SWOG (2.1-14.5 years) and in the differed arm, it was 12.3 years 8794 and EORTC 22911 convincingly showed that the (1.3-14.2; p: 0.0004). The data continue to support the primary risk of treatment failure was local, suggesting use of ADT in node-positive disease but it is unknown that the adjuvant studies treating these patients solely whether ADT improves overall survival in high-risk pa- with systemic therapy might have limited benefits. Addi- tients with negative lymph node. Mc Leod et al. [38] tionally, offering adjuvant irradiation to all patients with have recently published the preliminary results of a large pT3 disease could result in overtreatment for a number of trial evaluating the efficacy and the tolerability of bicatu- patients, as it was exemplified by the fact that in the ob- lamide (150 mg daily) as adjuvant therapy after PR or RT servation arm of the EORTC and SWOG studies, respec- in patients with locally advanced prostate cancer. A total tively 52.6% and 38% of patients did not show any bio- of 8.133 patients were recruited for this placebo-controlled chemical relapse. Therefore a better definition of high- double-blinded randomized study. With a median fol- risk groups is necessary to reduce the overtreatment rate low-up of 7.4 years, bicalutamide significantly reduced the of RT, side effects and care costs. risk of objective progression compared to placebo (HR: Wiegel et al. [43], in their preliminary evaluation of a 0.75; IC 95%: 0.61-0.91; p: 0.004). There was no statisti- Phase III study comparing RP followed by RT (60 Gy) cally significant difference between the two groups in with RP alone in patients with pT3 disease, have reported a terms of overall survival after RP (HR: 1.09; IC 95%: significant improvement of relapse-free survival among 0.85-1.39; p: 0.51). Again, in men with locally advanced patients receiving adjuvant therapy compared to the con- prostate cancer, by 5 years follow-up, the Study by the trol arm; particularly in patients with a preoperative PSA> Scandinavian Prostate Group suggests benefits in terms of 10 ng/ml, pT3b and Gleason 8 as well as positive margins. progression free survival (PFS) of adding bicalutamide to RP, RT or watchfull waiting [39]. In contrast, bicalutamide 3.2.3 Adjuvant RT and Andogen Deprivation provides no benefit in patients with localized prostate can- Therapy after RP RADICALS is a large international Phase III randomized cer but rather may decrease PFS. controlled trial addressing the RT to the tumoral bed (66 In addition, the optimal duration of the hormonal ther- Gy) and ADT after RP [44]. The first randomization, apy remains to be established. Adjuvant therapy has been performed within the 3 months after RP (the RT timing), used for duration of 2 years or 3 years in phase III trials consists in randomizing patients to immediate RT versus in men with high-risk disease and the question of adverse salvage RT. The second randomization is performed be- effects should be considered in this setting. fore giving RT (RT duration hormonal therapy) between 3.2.2 Adjuvant Radiotherapy (RT) no HT, short-term HT (6 months duration) and long-term Two large randomized trials, European Organization for HT (24 months). The primary end point will be the can- research and Treatment of cancer (EORTC) trial 22911 cer-specific survival (CCS), the secondary end point will and Southwest Oncology Group (SWOG) trial 8794, be overall survival. Especially, The RADICALS trial is

Copyright © 2010 SciRes JCT Adjuvant Treatment for High-Risk Operable Prostate Cancer 13 designed to identify treatment options that could achieve nisone added to the standard of care (i.e., surveillance an absolute increase in 10-years CCS of =>5%. 2600 with the addition of androgen deprivation at the time of patients in the RT timing randomization could detect an biochemical relapse) for patients, who are at high risk for increase in 10-years CCS from 70 to 75% with 80% relapse after RP. Patients “veterans”, are stratified for power, or from 80 to 85% with 90% power and 5% sig- PSA, Gleason score, tumor stage, and the presence of nificant level. 3500 patients would be required for the HT positive margins. A planned 636 patients will be accrued duration randomization. The patients who have a PSA and randomized to one of two treatment arms: docetaxel level after RP<0.2 ng/ml with some risk factors for dis- plus prednisone administred every 3 weeks for 18 weeks ease recurrence that is pT3, positive margins, Gleason or surveillance alone. Patients will then be followed for a score>6, pre-operative PSA level of >10 ng/ml or a com- minimum of 1 yr and a maximum of 5 yr. The study is bination of these criteria will be included. designed with 90% power to detect a reduction in the 5-year progression rate from 60% to 45% (15% absolute 3.2.4 Adjuvant Chemotherapy: difference, 25% relative difference). The estimated study In men with metastatic hormone-refractory Pca, two Phase Completion Date is June 2011. III trials have demonstrated that chemotherapy based on taxotere is superior to that based on mitoxantrone plus 3.2.5 Ajuvant Chemohormonal Therapy: prednisone [45,46]. Patients receiving docetaxel-based Recent neoadjuvant studies have indicated that combin- chemotherapy had a longer progression-free survival and a ing hormonal and chemotherapy is feasible and safe (for better quality of life. They improved overall survival (2 review see ref [52]). Although the impact of chemother- months benefit) as compared with mitoxantrone groups. apy on survival need to be proved in randomized trials, it Overall, the use of Docetaxel is now a standard for men is interesting to note that neoadjuvant studies, wherein with hormone refractory disease. Conversely, adjuvant hormonal therapy was not included, consistently showed chemotherapy is not standard for high risk patients after declines in preoperative PSA level, ranging from 20% to RP. Should adjuvant chemotherapy be administered? Is 60% after chemotherapy. This indicated the likelihood of chemotherapy the next step? [47] These questions need to an antitumoral effect of these drugs in high risk patients be addressed in specific trials. irrespective of hormonal treatment [53–56]. Schmidt et al. [48,49] from the National Prostate Can- In their study, Pummer et al. [57] evaluated whether cer Group randomly assigned 184 patients with localized patients with previously untreated advanced Pca benefit advanced prostate cancer to one of the three following from combining total ADT with weekly epirubicin che- arms: 2 years of oral cyclophosphamide, estramustine- motherapy Patients with either metastatic (n=117) or lo- phosphate for 2 years versus observation. After 10 years cally advanced (n=28) were randomly allocated to treat- of follow-up, the estramustine-phosphate group had an ment with ADT by bilateral orchiectomy and flutamide improvement in relapse-free survival but there was no 250 mg or ADT plus weekly epirubicin 25 mg/m2 i.v. for difference in overall survival. 18 weeks. At a median follow-up of 81 months, progres- In patients at high-risk for occult distant disease fol- sion-free survival and overall survival in the ADT and lowing RP, a phase II trial of adjuvant docetaxel was E-ADT groups were 12 and 18 months (p<0.02) and 22 performed [50]. Treatment consisted in 6 cycles of 35 and 30 months respectively (p=0.12). Subjective quality mg/m2 docetaxel weekly given from 4 to 12 weeks fol- of life assessment showed no impairment of quality of lowing RP. At a median follow-up of 29.2 months (range life by epirubicin treatment. Objective toxicities were 1.6 to 39.2), 26 of 46 evaluable patients (60.5%) relapsed. generally mild with either treatment. The authors con- The observed median PFS was 15.7 months (95% CI: cluded that the combination of ADT and epirubicin was 12.8-25.1). This PFS is longer than the normal predicted well tolerated by patients with advanced Pca and resulted 10 months for these patients, and adjuvant docetaxel had in a significant extension of progression-free survival. significant but acceptable toxicity in high-risk patients. Wang et al. [58] have randomly assigned 96 patients Grade III toxicity occurred in 20 patients (26%) including with clinical T3 or T4 disease or metastatic disease to dyspnea in 4, fatigue and cardiac arrhythmia in 3 and mitoxantrone plus combined anfrogen bockade versus diarrhea, dizziness, nausea, acute vascular leak syndrome combined androgen blockade alone. In the 38 patients and hyperglycemia in 2. The incidence of Grade IV tox- without metastatic disease, a higher initial objective re- icity was relatively low and appeared in 3 patients. Seven sponse (95% versus 53%; p=0.0008) and median survival patients died including 4 of prostate cancer, 1 with in- (80 versus 36 months; p=0.04) were observed in patients tra-abdominal bleeding during treatment and 2 of pneu- healed with mitoxantrone plus combined androgen abla- monia and sudden cardiac deaths following treatment. tion. The Veterans Affairs Cooperative Study 553 [51] has In a prospective randomized Phase II study (trial been designed to prospectively evaluate the efficacy of PR005) [59] of adjuvant paclitaxel and ADT versus ADT early adjuvant chemotherapy, using docetaxel and pred- alone, 47 patients with high-risk Pca were randomized

Copyright © 2010 SciRes JCT 14 Adjuvant Treatment for High-Risk Operable Prostate Cancer after PR between paclitaxel 100 mg/m2 once a week for 8 need in high-risk patients with Pca: weeks and ADT for 3 years versus ADT for 3 years. The 1) To select patients with high-risk of residual disease mean age was 58 year-old [51–67], the mean PSA con- after PR (pathological margin). centration and Gleason score were respectively 18ng/ ml In stage I or II breast cancer treated with breast-con- and 7.4 [7–9]. Toxicity, quality of life and functional serving surgery and RT, the patients with close margins results were compared between the two arms. With a and those with negative margins both have a rate of local mean follow up of 36 months, 23 patients receiving pa- recurrence (LR) of 7%. It is interesting to note that clitaxel and ADT were evaluated. Toxicity and side effets women with extensively positive margins have an LR of were assessed using the National Cancer Institute’s 27%, whereas patients with focally positive margins had Common Toxicity Criteria (version.2). Alopecia was an intermediate rate of LR of 14%. The use of systemic observed in 100% of the cases. No hematologic toxicity therapy adds significant effect on the risk-ratio of LR was noted. 4 patients had neurological disorders in fin- [61]. In Pca, data from the Mayo clinic indicate that 76% gers (86% of grade I), 2 patients had nausea and/or vom- of patients with no positive surgical margin and 65% of iting disorders (grade I and II), 2 had asthenia (grade I) patients with a single-positive margin after RP remain and one patient developed cardiac insufficiency not due biochemically and clinically free from disease by 5 years; to chemotherapy. One grade III febrile neutropenia was 62% with two or more positive margins had no evidence reported. These preliminary results indicated that adju- of disease by 5 years [62]. SWOG 8794 and EORTC vant paclitaxel-based chemotherapy associated with ADT 22911 showed evidence that the primary risk of treatment was a safe and well tolerated approach. failure was local. In recent reports of EORTC 22911 using the grading, staging and surgical margins status de- 4. Local Control and Systemic Therapy termined by a central pathology review, it was shown that The effect of RT on survival should be considered in the the margin status was a stronger predictor for the magni- context of systemic therapy, which is thought to be effec- tude of the treatment benefit [63,64]. Thus, while the tive against distant disease. In breast cancer, adjuvant 5-year biochemical PFS rates were 67.4% (95% CI: systemic therapy reduces the likelihood of both local and 56.1% to 76.3%) and 76.2% (95% CI: 66.1% to 83.6%) distant recurrence. A subgroup analysis in the EBCTCG for patients with negative margins in the control versus metanalysis of local therapy showed that the use of RT irradiation arm, they were 77.6% (95% CI: 68.8% to after mastectomy in node-positive patients improved 84.2%) and 48.5% (95% CI: 39.6% to 58.9 %) for the 15-year survival only in patients who also received adju- patients with positive surgical margin in the control ver- vant systemic therapy and not in patients who were sus irradiation arm [61]. Treatment failure is now docu- treated with mastectomy alone [3]. In high-risk patients mented to be primarily in the area of the prostate fossa, for distant metastases, such as women with positive and adjuvant radiation reduces both biochemical and lymph nodes, RT in the absence of systemic therapy can clinical local recurrence (22% versus 8%). In the obser- improve survival only in the rare patients with residual vation arm of the EORTC study, the rate of clinical local local disease who have no distant dissemination. In con- treatment failure was four times the rate of systemic fail- trast, in node-positive patients treated with mastectomy ure. In view of these data, patients with high-risk prostate and adjuvant systemic therapy, RT will potentially con- cancer after PR should be given adjuvant RT as standard tribute to survival in patients in whom systemic therapy treatment [42] however immediate postoperative RT eradicates microscopic metastases but not residual local might not be recommended for Pca patients with negative disease. surgical margins. What is also pertinent for this review is recent evi- 2) To identify the patients in whom the influence of dence that, in men experiencing an increasing PSA after local RT on mortality will be reduced and possibly elimi- their primary local treatments (RP or RT), docetaxel nated with systemic therapy. -based systemic therapy administred every three weeks In breast cancer, Marks et al. [65] suggest that benefits was able to reduce PSA level [60]. Authors reported a of local therapy on survival has an inverted-U-shaped or decreased > or =50% in 17 of 35 patients (48.5%) and > parabolic relationship with increasing effective systemic or =75% in seven of 35 patients (20%) with docetaxel. therapy, so that the survival benefit derived from better Again, this demonstrated the activity of chemotherapy local therapy increases with increasing effective therapy against prostate cancer cells. In this study, chemotherapy but only to a certain threshold of effectiveness and then (for up to 6 cycles) was followed by hormone therapy. In declines. In the SWOG study, in the observation group, five of 33 men, the PSA remains at 0.1 ng/mL at a me- the rate of clinical local failure was 24% versus 16% dis- dian of 18.9 months. Herein, it is further interesting to tant metastases. By improving the local control, adjuvant note that three of these five men had soft tissue metasta- RT was associated with a reduction in the proportion of sis at entry but remain in complete remission. patients with metastases (16 to 7%). Based on the above The influence of RT on local control emphazise the mentioned observations, these data also suggest that of-

Copyright © 2010 SciRes JCT Adjuvant Treatment for High-Risk Operable Prostate Cancer 15 fering these patients systemic therapy could lead to lim- The design was a 4 arms randomization to immediate ited improvement in survival [42]. The fact remains that versus delayed therapy with ADT plus or without che- reduction in local recurrence with systemic therapy with motherapy. This trial was designed to provide valuable and without RT must be reported in further studies. The information on both hormonal therapy and chemotherapy new EORTC trial 22043-30041 aims at recruiting 600 in the adjuvant setting. In particular, it would have helped patients with a PSA level of =<0.2 ng/ml after RP and to define the optimal timing in adjuvant therapy among testing the value of adding 6 months of ADT to RT. The immediate postoperative period and delayed when in- RTOG 0534 trial in patients with PSA level of 0.2-2 creasing PSA was detected [66]. Of interest is that an ng/ml after RP will compare treatment with adjuvant RT important number of men were accrued prior to closure, alone or with additional 5 months of ADT or similar and useful data on their outcomes might be possible. ADT with pelvic RT. A Phase II randomized trial PR005 aims at using paclitaxel as an adjuvant for high-risk disease (Table 3) 5. Clinical Studies with Adjuvant ADT [59]. All patients will receive 4 years of ADT using and/or Chemotherapy LH-RH agonist for three years and bicalutamide 50 mg for 1 month, and half will be treated with 8 weekly cycles The role of adjuvant therapy in the high-risk population of 100 mg/m2 paclitaxel. An approach to maximizing following RP needs systematic study. From experience in tumor-cell death with adjuvant chemotherapy is to use colon and breast cancer, active agents in the metastatic optimal doses of active chemotherapy drugs administered setting seem to be more beneficial when used in patients sequentially with a shortened scheduling interval. This with earlier stage disease. Prospects for long-term sur- approach, called “dose-dense,” increases dose in- vival following surgical treatment of localized breast tensity (drug delivery over time) by reducing the inter- cancer have significantly improved with the widespread treatment interval for chemotherapy delivery. A number use of adjuvant systemic chemotherapy. According to the of preclinical studies suggest that continuous dosing of uptaded 2005, Early Breast cancer Trialist collaborative chemotherapy with a very short interdose interval, so meta-analysis, combination chemotherapy was associated with an approximate 23% reduction in the risk of breast Table 1. SWOG 9921 adjuvant trial in patients with Pca at cancer recurrence. The greatest absolute benefit was no- high risk after RP ticed in younger women with lymph node–positive disease. In absolute terms, a 5-year course’ of a selective High-risk criteria Randomized adjuvant treatment oestrogen receptor modulator tamoxifen for patients with -Arm A: ADT 24 months oestrogen receptor-positive tumors reduces the annual T3a,+margin and Gleason 7 breast cancer death rate by 31% largely irrespective of chemotherapy [3]. T3 on biopsy N0M0 As with the use of adjuvant treatment in breast cancer, T3 b-T4 or pN1 -Arm B: ADT 24months three randomized studies were planned to investigate Or Gleason=>8 and mitoxantrone chemohormonal adjuvant setting in patients considered to PSA>20 ng/ml 12 mg: m2 d1+ be at high-risk Pca after RP. prednisone 5mg BID SWOG 9921 was intiated but it closed prematurely in January 2007 (Table 1). The primary endpoint of this N: 1380 (to detect D1-D21 Q 3 weeks x6 trial was initially overall survival. All patients should A 30% survival difference) have received 2 years of combined ADT using goserelin and bicalutamide, and half should have been treated with Table 2. TAX3501 adjuvant trial in patients with Pca at 6 cycles of 12 mg/m2 mitoxantrone plus 5 mg prednisone high risk after RP twice daily. RT to the prostate was allowed. Study accrual was held when three cases of leukemia (AML) were High-risk criteria Randomized adjuvant treatment reported in the mitoxantrone-containing arm. Therefore, -Arm A: observation all patients in the chemotherapy-containing arm stopped Kattan nomogram taking mitoxantrone. Although this phase III has stopped, stratification progression 2 years of ADT will be continued in active patients in 0-20% -ADT both arms and it may be possible to obtain precious information given the high number of patients included. 20-40 % -or ADT+docetaxel A second Phase III TAX 3501 aimed at using do- 40-60% cetaxel as an adjuvant treatment for high-risk disease -Arm B with an accrual goal of 2172 patients (Table 2). The pri- N: 1696 -ADT mary endpoint was PFS. Unfortunately, this study failed to meet its accrual target and was closed very recently. -ADT+docetaxel

Table 3. PROO5 adjuvant trial in patients with Pca at high markers, namely MUC1 and AZGP1, differentially ex- risk after RP pressed among indentified tumor subgroups. MUC1 and High-risk criteria Randomized adjuvant treatment AZGP1were found to be strong predictor of tumor recurrence. In Kaplan-Meier survival analysis, positive MUC1 Age =<67 years -Arm A: ADT 36 months staining was associated with significantly shorter time to cT2-3a N0M0 recurrence, while strong immunostaining of AZGP1 was pT3 b-pT4 and/or pN1 -Arm B: ADT 36 months weekly associated with significantly prolonged time to recur- Gleason=>7 paclitaxel 100 mg/m2 rence. Importantly, these genes were found, in multivari- PSA>10 ng/ml ate analysis, as additional prognostic information over and above the known risk factors of tumor grade, stage, N: 178 (to detect a and preoperative PSA. These genes also provided inde- 20% PFS difference) pendent prognostic value, suggesting that using two genes improves the accuracy of tumor subtyping and called “metronomic” scheduling, may enhance the anti- prognostication. angiogenic effects of chemotherapy in breast cancer Glinsky et al. [88] reported a Pca recurrence predictor [66–71]. This current study tests a sequential dose-dense algorithm that appeared suitable for stratification of pa- of weekly administration of paclitaxel (cumulative dose: tients at the time of diagnosis into poor-and good-prog- 800 mg/m2) for which the feasibility and efficacy has nosis subsets, this, with a statistically significant differ- been assessed previously [26,27]. In this study, frozen ence in the disease-free survival after RP. It could pro- prostate tissue will be obtained from men undergoing RP vide additional predictive value over conventional prog- who are enrolled in either the treatment or the control nostic factors such as PSA level and Gleason sum. Tom- arms of the trial. These samples will be analyzed for their kins et al. [89] reported that a majority of Pca exhibit mRNA level expression patterns in an attempt to draw up fusions between the control region of an androgen regu- outcome prediction models. Likewise, from these ar- lated gene TMPRSS2 and the coding region of the ETS ray-based methods of expression analysis, it would be family of transcription factors, most frequently ERG and ideal if we could predict the sensitivity to chemothera- much less frequently ETV1 and ETV4. The fusions are peutic agents and the response to chemotherapy. associated with an increased risk of cancer progression in 6. Microarrays Analysis patients treated surgically [90,91]. Recently, a four- variable model predictive of cancer-specific outcome Microarrays analysis has been used to characterize the incorporate gene expression of topoisomerase-2a, cad- molecular profiles of breast cancer [72–74]. Important herine-10, the fusion status based ERG, ETV1 and ETVA advances are being made in the use of genetic analysis to expression and the aneuploïd status in men with high- determine the risk of recurrence and to predict a tumor’s grade Pca treated with RP was established [92]. The trial responsiveness to adjuvant chemotherapy or tamoxifen in POO5 has used gene expression profiling to define sub- breast cancer [75–79]. These approaches may be infor- groups of high-risk Pca associated with good or poor mative to determine high-risk Pca for local recurrence or outcome. The refined gene-expression signature associ- distant recurrence. Given that initiation and progression ated with metastases contained three upregulated and of Pca involve multiple changes in gene expression, thirty down regulated gene (Personal communication). cDNA microarray technology has been recently used to Collectively, these data illustrate the potential helpful- identify disease-related gene expression patterns in pros- ness of expression profiling in the identification of tate samples [80–82] This approach has successfully de- high-risk patients as well as in the development of new tected alterations in several candidate genes associated biological markers and prognostic markers. with Pca progression [83,84]. However, there is not any definitive molecular classification that can consistently 7. Conclusions and reliably predict the clinical behaviour of Pca yet. Patients with high-risk Pca after RP should be offered Nevertheless, gene expression profiling offers an alterna- adjuvant RT as standard treatment. However, a policy of tive means to distinguish aggressive tumor biology and adjuvant RT would result in significant overtreatment: in may improve the accuracy of outcome prediction for pa- the observation arm of the EORTC and SWOG studies, tients with Pca treated by RP [14,85,86]. Interestingly, 52.6% and 38% of patients did not show any biochemical Lapointe et al. [87], so as to further characterize the relapse. Thus, a better definition of high-risk groups is clinical relevance of tumor subtypes identified from a warranted to reduce the overtreatment rate of RT and to gene expression profiling, used immunohistochemistry reduce the side effects and cost of this adjuvant treatment. on tissue microarrays in an independent set of 225 pros- An alternative approach to reduce the number of treated tate tumors in order to assess two genes as surrogate patients might be to identify subsets of patients who may

Copyright © 2010 SciRes JCT Adjuvant Treatment for High-Risk Operable Prostate Cancer 17 significantly benefit from immediate post operative RT. [9] S. J. Freedland, E. B. Humphreys, L. A. Mangold, et al., Further improved in local control with adjuvant systemic “Risk of prostate cancer-specific mortality following bio- treatment would also enable better results. Accumulating chemical recurrence after radical prostatectomy,” Journal clinical and preclinical data suggests that the use of early of the American Medical Association, Vol. 294, pp. 433– HT will improve the outcome in patients with high-risk 439, 2005. localized Pca. Yet, two randomized Phase III trials [10] V. Ravery, C. Chastang, M. Toublanc, L. Boccon-Gibod, evaluating the effect of adjuvant hormonal therapy with V. Delmas, and L. Boccon-Gibod, “Percentage of cancer or without chemotherapy in high-risk patients after RP on biopsy cores accurately predicts extracapsular exten- have been prematurely closed. Another problem is to sion and biochemical relapse after radical prostatectomy for T1-T2 prostate cancer,” European Urology, Vol. 37, better distinguish diseases that will be cured with local pp. 449–455, 2000. treatement only, from those that will require an adjuvant approach. Gene expression profiling of Pca offers an al- [11] A. V. D'Amico, M. H. Chen, K. A. Roehl, and W. J. Cata- lona, “Preoperative PSA velocity and the risk of death ternative means to distinguish aggressive tumor biology from prostate cancer after radical prostatectomy,” New and may improve the accuracy of outcome prediction for England Journal of Medicine, Vol. 351, pp. 125–135, patients with Pca treated by RP. 2004. [12] M. W. Kattan, J. A. Eastham, A. M. Stapleton, T. M. REFERENCES Wheeler, and P. T. Scardino, “A preoperative nomogram [1] A. V. D'Amico, R. Whittington, S. B. Malkowicz, et al., for disease recurrence following radical prostatectomy for “Biochemical outcome after radical prostatectomy, exter- prostate cancer,” Journal of the National Cancer Institute, nal beam radiation therapy, or interstitial radiation therapy Vol. 90, pp. 766–771, 1998. for clinically localized prostate cancer,” Journal of the [13] M. W. Kattan, T. M. Wheeler, and P. T. Scardino, “Post- American Medical Association, Vol. 280, pp. 969–974, operative nomogram for disease recurrence after radical 1998. prostatectomy for prostate cancer,” Journal of Clinical [2] “Adjuvant systemic therapy for women with node-posi- Oncology, Vol. 17, pp. 1499–1507, 1999. tive breast cancer,” The Steering Committee on Clinical [14] D. Weckermann, P. Muller, F. Wawroschek, G. Krawczak, Practice Guidelines for the Care and Treatment of Breast G. Riethmuller, and G. Schlimok, “Micrometastases of Cancer, Canadian Medical Association Journal, Vol. 158, bone marrow in localized prostate cancer: Correlation Supplement 3, pp. S52–64, 1998. with established risk factors,” Journal of Clinical Oncol- [3] “Effects of chemotherapy and hormonal therapy for early ogy, Vol. 17, pp. 3438–3443, 1999. breast cancer on recurrence and 15-year survival: An over- [15] P. B. Morgan, A. L. Hanlon, E. M. Horwitz, M. K. Buy- view of the randomized trials,” Early Breast Cancer Trial- younouski, R. G. Uzzo, and A. Pollack, “Timing of bio- ists’ Collaborative Group, Lancet, Vol. 365, pp. 1687–1717, chemical failure and distant metastatic disease for low-, 2005. intermediate, and high-risk prostate cancer after radio- [4] C. G. Moertel, T. R. Fleming, J. S. Macdonald, et al., therapy,” Cancer, Vol. 110, pp. 68–80, 2007. “Levamisole and fluorouracil for adjuvant therapy of re- [16] H. Miyake, T. Kurahashi, I. Hara, A. Takenaka, and M. sected colon carcinoma,” New England Journal of Medi- Fujisawa, “Significance of micrometastases in pelvic cine, Vol. 322, pp. 352–358, 1990. lymph nodes detected by real-time reverse transcriptase [5] T. Andre, C. Boni, L. Mounedji-Boudiaf, et al., “Ox- polymerase chain reaction in patients with clinically lo- aliplatin, fluorouracil, and leucovorin as adjuvant treat- calized prostate cancer undergoing radical prostatectomy ment for colon cancer,” New England Journal of Medi- after neoadjuvant hormonal therapy,” British Journal of Urology International, Vol. 99, pp. 315–320, 2007. cine, Vol. 350, pp. 2343–2351, 2004. [17] M. Fukuda, M. Egawa, T. Imao, H. Takashima, K. Yoko- [6] C. Twelves, A. Wong, M. P. Nowacki, et al., “Capecit- abine as adjuvant treatment for stage III colon cancer,” yama, and M. Namiki, “Detection of sentinel node mi- New England Journal of Medicine, Vol. 352, pp. 2696– crometastasis by step section and immunohistochemistry 2704, 2005. in patients with prostate cancer,” Journal of Urology, Vol. 177, pp. 1313–1317, discussion 1317, 2007. [7] D. P. Lubeck, M. S. Litwin, J. M. Henning, et al., “The CaPSURE database: A methodology for clinical practice [18] S. F. Shariat, M. P. Roudier, G. E. Wilcox, et al., “Com- and research in prostate cancer,” CaPSURE Research parison of immunohistochemistry with reverse transcrip- Panel, Cancer of the Prostate Strategic Urologic Research tion-PCR for the detection of micrometastatic prostate Endeavor, Urology, Vol. 48, pp. 773–777, 1996. cancer in lymph nodes,” Cancer Research, Vol. 63, pp. 4662–4670, 2003. [8] C. R. Pound, A. W. Partin, M. A. Eisenberger, D. W. Chan, J. D. Pearson, and P. C. Walsh, “Natural history of [19] A. J. Stephenson, A. Smith, M. W. Kattan, et al., “Inte- progression after PSA elevation following radical prosta- gration of gene expression profiling and clinical variables tectomy,” Journal of the American Medical Association, to predict prostate carcinoma recurrence after radical Vol. 281, pp. 1591–1597, 1999. prostatectomy,” Cancer, Vol. 104, pp. 290–298, 2005.

[20] M. Ohori, J. R. Goad, T. M. Wheeler, J. A. Eastham, T. C. Smeets, and J. Wils, “Phase II study of epirubicin in ad- Thompson, and P. T. Scardino, “Can radical prostatec- vanced hormone-resistant prostatic carcinoma,” British tomy alter the progression of poorly differentiated pros- Journal of Urology, Vol. 70, pp. 641–642, 1992. tate cancer?” Journal of Urology, Vol. 152, pp. 1843– [33] R. Petrioli, A. I. Fiaschi, D. Pozzessere, et al., “Weekly 1849, 1994. epirubicin in patients with hormone-resistant prostate [21] H. Van Poppel, H. Goethuys, P. Callewaert, L. Vanuytsel, cancer,” British Journal of Cancer, Vol. 87, pp. 720–725, W. Van de Voorde, and L. Baert, “Radical prostatectomy 2002. can provide a cure for well-selected clinical stage T3 [34] J. T. Isaacs, “The timing of androgen ablation therapy prostate cancer,” European Urology, Vol. 38, pp. 372– and/or chemotherapy in the treatment of prostatic cancer,” 379, 2000. Prostate, Vol. 5, pp. 1–17, 1984. [22] N. Craft, C. Chhor, C. Tran, et al., “Evidence for clonal [35] B. J. Eigl, S. E. Eggener, J. Baybik, et al., “Timing is outgrowth of androgen-independent prostate cancer cells everything: Preclinical evidence supporting simultaneous from androgen-dependent tumors through a two-step rather than sequential chemohormonal therapy for prostate process,” Cancer Research, Vol. 59, pp. 5030–5036, cancer,” Clinical Cancer Research, Vol. 11, pp. 4905– 1999. 4911, 2005. [23] J. T. Isaacs, “Relationship between tumor size and cura- [36] E. M. Messing, J. Manola, M. Sarosdy, G. Wilding, E. D. bility of prostatic cancer by combined chemo-hormonal Crawford, and D. Trump, “Immediate hormonal therapy therapy in rats,” Cancer Research, Vol. 49, pp. 6290– compared with observation after radical prostatectomy 6294, 1989. and pelvic lymphadenectomy in men with node-positive [24] D. Friedland, J. Cohen, and R. Miller, Jr., et al., “A phase prostate cancer,” New England Journal of Medicine, Vol. II trial of docetaxel (Taxotere) in hormone-refractory 341, pp. 1781–1788, 1999. prostate cancer: Correlation of antitumor effect to phos- [37] E. M. Messing, J. Manola, J. Yao, et al., “Immediate ver- phorylation of Bcl-2,” Seminars in Oncology, Vol. 26, pp. sus deferred androgen deprivation treatment in patients 19–23, 1999. with node-positive prostate cancer after radical prostatec- [25] W. Berry, S. Dakhil, M. A. Gregurich, and L. Asmar, tomy and pelvic lymphadenectomy,” Lancet Oncology, “Phase II trial of single-agent weekly docetaxel in hor- Vol. 7, pp. 472–479, 2006. mone-refractory, symptomatic, metastatic carcinoma of [38] D. G. McLeod, P. Iversen, W. A. See, T. Morris, J. Arm- the prostate,” Seminars in Oncology, Vol. 28, pp. 8–15, strong, and M. P. Wirth, “Bicalutamide 150 mg plus stan- 2001. dard care vs standard care alone for early prostate cancer,” [26] B. J. Roth, B. Y. Yeap, G. Wilding, B. Kasimis, D. British Journal of Urology International, Vol. 97, pp. McLeod, and P. J. Loehrer, “Taxol in advanced, hor- 247–254, 2006. mone-refractory carcinoma of the prostate. A phase II trial [39] P. Iversen, J. E. Johansson, P. Lodding, et al., “Bicalu- of the Eastern Cooperative Oncology Group,” Cancer, tamide (150 mg) versus placebo as immediate therapy Vol. 72, pp. 2457–2460, 1993. alone or as adjuvant to therapy with curative intent for [27] C. Trivedi, B. Redman, L. E. Flaherty, et al., “Weekly early nonmetastatic prostate cancer: 5.3-year median fol- 1-hour infusion of paclitaxel. Clinical feasibility and effi- lowup from the Scandinavian Prostate Cancer Group cacy in patients with hormone-refractory prostate carci- Study Number 6,” Journal of Urology, Vol. 172, pp. noma,” Cancer, Vol. 89, pp. 431–436, 2000. 1871–1876, 2004. [28] I. F. Tannock, D. Osoba, M. R. Stockler, et al., “Chemo- [40] M. Bolla, H. van Poppel, L. Collette, et al., “Postoperative therapy with mitoxantrone plus prednisone or prednisone radiotherapy after radical prostatectomy: A randomised alone for symptomatic hormone-resistant prostate cancer: controlled trial (EORTC trial 22911),” Lancet, Vol. 366, A Canadian randomized trial with palliative end points,” pp. 572–578, 2005. Journal of Clinical Oncology, Vol. 14, pp. 1756–1764, [41] I. M. Thompson, Jr., C. M. Tangen, J. Paradelo, et al., 1996. “Adjuvant radiotherapy for pathologically advanced pros- [29] P. W. Kantoff, S. Halabi, M. Conaway, et al., “Hydrocor- tate cancer: a randomized clinical trial,” Journal of the tisone with or without mitoxantrone in men with hor- American Medical Association, Vol. 296, pp. 2329–2335, mone-refractory prostate cancer: Results of the cancer and 2006. leukemia group B 9182 study,” Journal of Clinical On- [42] G. P. Swanson, M. A. Hussey, C. M. Tangen, et al., cology, Vol. 17, pp. 2506–2513, 1999. “Predominant treatment failure in postprostatectomy pa- [30] R. Morant, S. F. Hsu Schmitz, J. Bernhard, et al., “Vi- tients is local: Analysis of patterns of treatment failure in norelbine in androgen-independent metastatic prostatic SWOG 8794,” Journal of Clinic Oncology, Vol. 25, pp. carcinoma–a phase II study,” European Journal of Cancer, 2225–2229, 2007. Vol. 38, pp. 1626–1632, 2002. [43] T. Wiegel, O. Bottke, N. Willich, et al., “Phase III results [31] S. Oudard, A. Caty, Y. Humblet, et al., “Phase II study of of adjuvant radiotherapy versus wait-and-see in patients vinorelbine in patients with androgen-independent prostate with pT3 prostate cancer following radical prostatectomy cancer,” Annals of Oncology, Vol. 12, pp. 847–852, 2001. (RO 96-02/AUO AP 09/05),” Journal of Clinic Oncology, [32] K. P. Delaere, H. Leliefeld, F. Peulen, E. W. Stapper, J. Annual Meeting Proceedings Part 1, Vol. 25, 18S (June

20, Supplement), pp. 5060, 2007. vant docetaxel before radical prostatectomy in patients [44] C. Parker, M. R. Sydes, C. Catton, et al., “Radiotherapy with high-risk localized prostate cancer,” Clinic Cancer and androgen deprivation in combination after local sur- Research, Vol. 11, pp. 5233–5240, 2005. gery (RADICALS): A new Medical Research Coun- [57] K. Pummer, M. Lehnert, H. Stettner, and G. Hubmer, cil/National Cancer Institute of Canada phase III trial of “Randomized comparison of total androgen blockade adjuvant treatment after radical prostatectomy,” British alone versus combined with weekly epirubicin in ad- Journal of Urology International, Vol. 99, pp. 1376–1379, vanced prostate cancer,” European Urology, Vol. 32, Sup- 2007. plement 3, pp. 81–85, 1997. [45] D. P. Petrylak, C. M. Tangen, M. H. Hussain, et al., “Do- [58] J. Wang, S. Halford, A. Rigg, R. Roylance, M. Lynch, cetaxel and estramustine compared with mitoxantrone and and J. Waxman, “Adjuvant mitozantrone chemotherapy in prednisone for advanced refractory prostate cancer,” New advanced prostate cancer,” British Journal of Urology In- England Journal of Medicine, Vol. 351, pp. 1513–1520, ternational, Vol. 86, pp. 675–680, 2000. 2004. [59] G. Ploussard, B. Paule, L. Salomon, Y. Allory, S. Terry, [46] I. F. Tannock, R. de Wit, W. R. Berry, et al., “Docetaxel D. Vordos, A. Hoznek, F. Vacherot, C. C. Abbou, S. Cu- plus prednisone or mitoxantrone plus prednisone for ad- line, and A. de la Taille, “Pilot trial of adjuvant paclitaxel vanced prostate cancer,” New England Journal of Medi- plus androgen deprivation for patients with high-risk cine, Vol. 351, pp. 1502–1512, 2004. prostate cancer after radical prostatectomy: Results on toxicity, side effects and quality-of-life,” Prostate Cancer [47] W. K. Oh and P. W. Kantoff, “Treatment of locally ad- and Prostatic Diseases, pp. 1–5, 2009. vanced prostate cancer: Is chemotherapy the next step?” Journal of Clinic Oncology, Vol. 17, pp. 3664–3675, 1999. [60] A. Hussain, N. Dawson, P. Amin, et al., “Docetaxel followed by hormone therapy in men experiencing increas- [48] J. D. Schmidt, R. P. Gibbons, G. P. Murphy, and A. Bar- ing prostate-specific antigen after primary local treatments tolucci, “Adjuvant therapy for clinical localized prostate for prostate cancer,” Journal of Clinical Orthodontics, Vol. cancer treated with surgery or irradiation,” European 23, pp. 2789–2796, 2005. Urology, Vol. 29, pp. 425–433, 1996. [61] C. C. Park, M. Mitsumori, A. Nixon, et al., “Outcome at 8 [49] J. D. Schmidt, R. P. Gibbons, G. P. Murphy, and A. Bar- years after breast-conserving surgery and radiation ther- tolucci, “Evaluation of adjuvant estramustine phosphate, apy for invasive breast cancer: Influence of margin status cyclophosphamide, and observation only for node-posi- and systemic therapy on local recurrence,” Journal of tive patients following radical prostatectomy and defini- Clinical Orthodontics, Vol. 18, pp. 1668–1675, 2000. tive irradiation,” Investigators of the National Prostate Cancer Project, Prostate, Vol. 28, pp. 51–57, 1996. [62] S. J. Kausik, M. L. Blute, T. J. Sebo, et al., “Prognostic significance of positive surgical margins in patients with [50] A. S. Kibel, E. Rosenbaum, M. W. Kattan, et al., “Adju- extraprostatic carcinoma after radical prostatectomy,” vant weekly docetaxel for patients with high risk prostate Cancer, Vol. 95, pp. 1215–1219, 2002. cancer after radical prostatectomy: A multiinstitutional pilot study,” Journal of Urology, Vol. 177, pp. 1777–1781, [63] T. H. Van der Kwast, L. Collette, and M. Bolla, “Adju- 2007. vant radiotherapy after surgery for pathologically advanced prostate cancer,” Journal of Clinical Orthodontics, [51] U. S. National Insititutes of Health Reference CSP #553 Vol. 25, pp. 5671–5672, 2007. chemotherapy after prostatectomy for high risk prostate carcinoma. Available from URL: http://www. clinicaltri- [64] T. H. Van der Kwast, M. Bolla, H. Van Poppel, et al., als. gov. “Identification of patients with prostate cancer who benefit from immediate postoperative radiotherapy: EORTC [52] M. Gleave and W. K. Kelly, “High-risk localized prostate 22911,” Journal of Clinical Orthodontics, Vol. 25, pp. cancer: A case for early chemotherapy,” Journal of Clinic 4178–4186, 2007. Oncology, Vol. 23, pp. 8186–8191, 2005. [65] L. B. Marks and L. R. Prosnitz, “Postoperative radiother- [53] R. Dreicer, C. Magi-Galluzzi, M. Zhou, et al., “Phase II apy for lung cancer: The breast cancer story all over trial of neoadjuvant docetaxel before radical prostatec- again?” International Journal of Radiation Oncology Bi- tomy for locally advanced prostate cancer,” Urology, Vol. ology Physics, Vol. 48, pp. 625–627, 2000. 63, pp. 1138–1142, 2004. [66] I. C. Henderson, D. A. Berry, G. D. Demetri, et al., “Im- [54] T. M. Beer, M. Garzotto, B. A. Lowe, et al., “Phase I proved outcomes from adding sequential Paclitaxel but study of weekly mitoxantrone and docetaxel before not from escalating Doxorubicin dose in an adjuvant prostatectomy in patients with high-risk localized prostate chemotherapy regimen for patients with node-positive cancer,” Clinic Cancer Research, Vol. 10, pp. 1306–1311, primary breast cancer,” Journal of Clinical Orthodontics, 2004. Vol. 21, pp. 976–983, 2003. [55] W. K. Oh, D. J. George, D. S. Kaufman, et al., “Neoad- [67] M. L. Citron, D. A. Berry, C. Cirrincione, et al., “Ran- juvant docetaxel followed by radical prostatectomy in pa- domized trial of dose-dense versus conventionally sched- tients with high-risk localized prostate cancer: A prelimi- uled and sequential versus concurrent combination che- nary report,” Seminars Oncology, Vol. 28, pp. 40–44, motherapy as postoperative adjuvant treatment of node- 2001. positive primary breast cancer: First report of Intergroup [56] P. G. Febbo, J. P. Richie, D. J. George, et al., “Neoadju- Trial C9741/Cancer and Leukemia Group B Trial 9741,”

Journal of Clinical Orthodontics, Vol. 21, pp. 1431–1439, [81] G. V. Glinsky, A. B. Glinskii, A. J. Stephenson, R. M. 2003. Hoffman, and W. L. Gerald, “Gene expression profiling [68] Y. Shaked, U. Emmenegger, G. Francia, et al., “Low-dose predicts clinical outcome of prostate cancer,” Joint Com- metronomic combined with intermittent bolus-dose mission International, Vol. 113, pp. 913–923, 2004. cyclophosphamide is an effective long-term chemother- [82] D. Singh, P. G. Febbo, K. Ross, et al., “Gene expression apy treatment strategy,” Cancer Research, Vol. 65, pp. correlates of clinical prostate cancer behavior,” Cancer 7045–7051, 2005. Cell, Vol. 1, pp. 203–209, 2002. [69] J. A. Sparano, M. Wang, S. Martino, et al., “Phase III [83] G. K. Reddy and S. P. Balk, “Clinical utility of microar- study of AC followed by paclitaxelor docetaxel given ray-derived genetic signatures in predicting outcomes in every 3 weeks or weekly in patients with axillary node- prostate cancer,” Clinical Genitourinary Cancer, Vol. 5, positive or high-risk node-negative breast cancer: Results pp. 187–189, 2006. of intergroup trial E1199,” Presented at the 28th Annual [84] S. M. Henshall, D. E. Afar, J. Hiller, et al., “Survival analy- San Antonio Breast Cancer Symposium, San Antonio, TX, sis of genome-wide gene expression profiles of prostate 8-11 December 2005 (abstract #48). cancers identifies new prognostic targets of disease re- [70] G. Bocci, K. C. Nicolaou, and R. S. Kerbel, “Protracted lapse,” Cancer Reseach, Vol. 63, pp. 4196–4203, 2003. low-dose effects on human endothelial cell proliferation [85] S. Bettuzzi, M. Scaltriti, A. Caporali, et al., “Successful and survival in vitro reveal a selective antiangiogenic prediction of prostate cancer recurrence by gene profiling window for various chemotherapeutic drugs,” Cancer Re- in combination with clinical data: A 5-year follow-up search, Vol. 62, pp. 6938–6943, 2002. study,” Cancer Research, Vol. 63, pp. 3469–3472, 2003. [71] H. Kaur and G. T. Budd, “Metronomic therapy for breast [86] A. Latil, I. Bieche, L. Chene, et al., “Gene expression cancer,” Current Oncology Reports, Vol. 6, pp. 49–52, 2004. profiling in clinically localized prostate cancer: A [72] T. Sorlie, C. M. Perou, R. Tibshirani, et al., “Gene ex- four-gene expression model predicts clinical behavior,” pression patterns of breast carcinomas distinguish tumor Clinical Cancer Research, Vol. 9, pp. 5477–5485, 2003. subclasses with clinical implications,” Proceedings of the [87] J. Lapointe, C. Li, J. P. Higgins, et al., “Gene expression National Academy of Sciences, USA, Vol. 98, pp. 10869– profiling identifies clinically relevant subtypes of prostate 10874, 2001. cancer,” Proceedings of the National Academy of Sci- [73] L. J. van't Veer, H. Dai, M. J. van de Vijver, et al., “Gene ences, USA, Vol. 101, pp. 811–816, 2004. expression profiling predicts clinical outcome of breast [88] G. V. Glinsky, A. B. Glinsky, A. J. Stephenson, A. J. cancer,” Nature, Vol. 415, pp. 530–536, 2002. Hoffman, R. M. Hoffman, and W. L. Gerald, “Gene ex- [74] M. J. van de Vijver, Y. D. He, L. J. van't Veer, et al., “A pression profiling predicts clinical outcome of prostate gene-expression signature as a predictor of survival in cancerf,” Joint Commission International, Vol. 113, pp. breast cancer,” New England Journal of Medicine, Vol. 913–923, 2004. 347, pp. 1999–2009, 2002. [89] S. A. Tomlins, D. R. Rhodes, S. Perner, S. M. Dha- [75] J. C. Chang, E. C. Wooten, A. Tsimelzon, et al., “Gene nasekaran, R. Mehra, X. W. Sun, S. Varambally, X. Cao, expression profiling for the prediction of therapeutic re- J. Tchinda, R. Kuefer, C. Lee, J. E. Montie, R. B. Shah, K. sponse to docetaxel in patients with breast cancer,” Lancet, J. Pienta, M. A. Rubin, and A. M. Chinnaiyan, “Recurrent Vol. 362, pp. 362–369, 2003. fusion of TMPRSS2 and ETS transcription factor genes in [76] J. C. Chang, E. C. Wooten, A. Tsimelzon, et al., “Patterns prostate cancer,” Science, Vol. 310, pp. 644–648, 2005. of resistance and incomplete response to docetaxel by [90] F. Demichelis, K. Fall, S. Perner, O. Andrén, F. Schmidt, gene expression profiling in breast cancer patients,” Jour- S. R. Setlur, Y. Hoshida, J. M. Mosquera, Y. Pawitan, C. nal of Clinical Orthodontics, Vol. 23, pp. 1169–1177, 2005. Lee, H. O. Adami, L. A. Mucci, P. W. Kantoff, S. O. [77] K. Iwao-Koizumi, R. Matoba, N. Ueno, et al., “Prediction Andersson, A. M. Chinnaiyan, J. E. Johansson, and M. A. of docetaxel response in human breast cancer by gene ex- Rubin, “TMPRSS2: ERG gene fusion associated with le- pression profiling,” Journal of Clinical Orthodontics, Vol. thal prostate cancer in a watchful waiting cohort,” Onco- 23, pp. 422–431, 2005. gene, Vol. 26, pp. 4596–4599, 2007. [78] M. P. Jansen, J. A. Foekens, I. L. van Staveren, et al., [91] M. Macaluso and A. Giordano, “TMPRSS2: ERG gene “Molecular classification of tamoxifen-resistant breast fusion a new genetic marker for prostate cancer progres- carcinomas by gene expression profiling,” Journal of sion,” Cancer Biology and Therapy, Vol. 6, pp. 46–47, 2007. Clinical Orthodontics, Vol. 23, pp. 732–740, 2005. [92] J. C. Cheville, R. J. Karnes, T. M. Therneau, F. Kosari, J. [79] M. A. Rubin and A. M. De Marzo, “Molecular genetics of M. Munz, L. Tillmans, E. Basal, L. J. Rangel, E. human prostate cancer,” Modern Pathology, Vol. 17, pp. Bergstralh, I. V. Kovtun, C. D. Savci-Heijink, E. W. Klee, 380–388, 2004. and G. Vasmatzis, “Gene panel model predictive of outcome in men at high-risk of systemic progression and [80] S. M. Dhanasekaran, T. R. Barrette, D. Ghosh, et al., death from prostate cancer after radical retropubic pro- “Delineation of prognostic biomarkers in prostate cancer,” statectomy,” Journal of Clinical Orthodontics, Vol. 20, No. Nature, Vol. 412, pp. 822–826, 2001. 26, pp. 3930–3936, 2008.

Novel Ethyl 2-(1-aminocyclobutyl)-5-(benzoyloxy)- 6-hydroxy-pyrimidine-4-carboxylate Derivatives: Synthesis and Anticancer Activities

D. Asha1, C. V. Kavitha2, S. Chandrappa1, D. S. Prasanna1, K. Vinaya1, Sathees C. Raghavan2, K. S. Rangappa1

1Department of Studies in Chemistry, University of Mysore, Manasagangotri, Mysore, India; 2Department of Biochemistry, Indian Institute of Science, Bangalore, India. Email: [email protected]; [email protected]

Received December 26th, 2009; revised February 2nd, 2010; accepted February 12th, 2010.

ABSTRACT To explore the anticancer activity of 2, 4, 5, 6-substituted pyrimidines, several ethyl 2-(1-aminocyclobutyl)-5-(benzoyloxy)-6-hydroxy-pyrimidine-4-carboxylate derivatives associated with the different substituted aromatic/aliphatic carboxamides and sulfonamides were synthesized. Different groups and position on phenyl ring attached to the carboxamide and sulfonamide of the pyrimidine led to two set of compounds. Their chemical structures were confirmed by IR, 1H NMR and LC/MS analysis. Cytotoxicity of all the synthesized compounds were examined on human leukemia cell lines (K562 and CEM). The preliminary results showed most of the derivatives exhibited good antitumor activity. Com- pound with para chloro substitution among carboxamides and compound with meta dichloro substitution among sul- phonamides exhibited significant antitumor activity with IC50 value of 14.0 µM and 15.0 µM respectively against K562 cell line. For comparison among electron donating groups between carboxamides and sulfonamides, compounds with para tert-butyl substitution were chosen for further studies. Cell cycle analysis suggests that both tert-butyl substituted compounds are able to induce apoptosis.

Keywords: Pyrimidine Derivatives, Cytotoxicity, Apoptosis, Leukemia, Cell Cycle Analysis

1. Introduction used as key building blocks for pharmaceutical agents. Pyrimidines have a long and distinguished history ex- Cancer is a terrible disease which is the leading death tending from the days of their discovery as important of the human population in some areas of the world. It constituents of nucleic acids to their current use in the is the second leading cause of death, behind cardio- chemotherapy of cancer and AIDS. The most important vascular disease, in the United States [1]. Cancer may pyrimidine derivatives are those upon which biological affect people at all ages, even fetuses, but the risk for organisms depend. Pyrimidine derivatives possessing most varieties increases with age. Leukemia is a cancer anticancer activity have been reported in the literature of the blood-forming cells. Leukemia is the most com- [3–9]. Pyrimidine derivatives comprise a diverse and mon childhood cancer, affecting more than 3,500 child- interesting group of drugs [2]. The subject has been ren in the United States every year. At present, there discussed recently [10]. Earlier, a comprehensive re- are three main methods of cancer treatment: surgery, view concerning pyrimidines has been published by radiation therapy, and chemotherapy. With the devel- Brown [11]. Pyrimidines in general are extremely im- opment of molecular biology, chemotherapy is becom- portant for their biological activities. In addition to ing a more important therapeutic method. Therefore, above-mentioned activities, pyrimidine derivatives designing new anticancer drugs with high-efficiency possessing analgesic [12], antileishmanial [13], antim- and broad-spectrum activity is a significant study area icrobial [14], antifungal [15], and anti-infective have today. also been reported in the literature. Many other exam- Heterocycles are ubiquitous to among pharmacy- ples of pyrimidine-based derivatives have been inves- eutical compounds [2]. Pyrimidine moiety is an impor- tigated as potential antitumor agents, including 2-phe- tant class of nitrogen containing heterocycles widely nylamino derivatives [16–19], 4-phenylamino deriva-

Copyright © 2010 SciRes JCT 22 Novel Ethyl 2-(1-aminocyclobutyl)-5-(benzoyloxy)-6-hydroxy-pyrimidine-4-carboxylate Derivatives: Synthesis and Anticancer Activities tives [20,21], 2.4-bis(phenylamino) derivatives [22], Pd/C in ethyl acetate. 4.6-bis(phenylamino) derivatives [23,24], and 4-aryl- substituted derivatives [25]. In addition, the pyrimidine 2.1.2 General Procedure for the Synthesis of Ethyl pharmacophore is found in several fused-ring examples 2(1-aminocyclobutyl)-5-(benzoyloxy)-6-hydroxy- of ATP-competitive protein kinase inhibitors, such as pyrimidine-4-carboxylate Derivatives 2(a-i) purines, quinazolines and pyrido-, pyrimido-, pyrazolo-, A solution of ethyl 2-(1-aminocyclobutyl)-5-(benzoy- and pyrrolo-pyrimidines [26–28]. In continuation of loxy)-6-hydroxypyrimidine-4-carboxylate (1) (1.0 eq) in o our efforts to get novel and potent heterocyclic anti- dichloromethane was taken and cooled to 0-5 C in ice leukemic agents [29–32], a number of novel pyrimidine bath. Triethyl amine (3.0 eq) was added to the cold mix- derivatives were synthesized and evaluated for their ture and stirred for 10 min and respective aryl carbonyl anticancer activity which we wish to report in this pa- chlorides (1.0 eq) were added, the mixture was stirred per. and allowed at room temperature for 5 hr. The progress of the reaction was monitored by TLC. Upon completion, 2. Methods water was added to reaction mixture and extracted with 2.1 Chemistry ethyl acetate. The organic layer was washed with 10% ammonium chloride solution followed by water wash and Infrared (IR) spectra were recorded using a Jasco dried with anhydrous sodium sulphate. The solvent was FTIR-4100 spectrometer in the wave number range of -1 1 evaporated and the crude product obtained was purified by 4000-400 cm . Nuclear magnetic resonance ( H NMR) column chromatography over silica gel (60-120 mesh) spectra were recorded on a Bruker AM 400 MHz spec- using dichloromethane: methanol (9:1) as eluent. The IR, trometer using DMSO-d6 as solvent and tetramethylsi- 1H NMR and mass spectroscopic data are given in Table 1. lane as an internal standard. The chemical shifts are expressed in  and the following abbreviations are used: s 2.1.3 Synthesis of Ethyl 2-(1-(3,5-dinitrobenzamido) (singlet), d (doublet), t (triplet), q (quartet) and m (multi- Cyclobutyl)-5-(benzoyloxy)-6-hydroxy- plet). Mass and purity were recorded on a LC-MSD- pyrimidine-4-carboxylate (2a) Trap-XCT. The purity of the compounds was checked by The product obtained was white solid from ethyl 2-(1- thin layer chromatography (TLC). Silica gel column aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimidine- chromatography was performed using Merck 7734 silica 4-carboxylate 1 (0.1 g, 0.28 mmol), 3, 5-dinitrobenzoyl gel (60-120 mesh) and Merck made TLC plates. All the chloride (0.064 g, 0.28 mmol) and triethylamine (0.085 g, reagents and chemicals were from Sigma Aldrich Chemi- 0.84 mmol). cals Pvt Ltd. 2.1.4 Synthesis of Ethyl 2-(1-(3-methoxybenzamido) Cyclobutyl)-5-(benzoyloxy)-6-hydroxy- 2.2.1 Synthesis of Ethyl 2-(1-aminocyclobutyl)-5- pyrimidine-4-carboxylate (2b) (benzoyloxy)-6-hydroxypyrimidine- The product obtained was white solid from ethyl 2-(1- 4-carboxylate (1) aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimidine- For the synthesis of the target key intermediate 1 the re- 4-carboxylate 1 (0.1 g, 0.28 mmol), 3-methoxy-benzoyl action sequences as reported in our previous article were chloride (0.048 g, 0.28 mmol) and triethylamine (0.085 g, followed [33]. The synthesis of compounds 3(a-i) is also 0.84 mmol). reported earlier [33]. Treating cyclobutanone with ammonium chloride and sodium cyanide in methanol gave 2.1.5 Synthesis of Ethyl 2-(1-(2-fluorobenzamido) 1-isocyanocyclobutanamine. Amine group of 1-isocya- Cyclobutyl)-5-(benzoyloxy)-6-hydroxy- nocyclobutanamine was protected by using benzyl chlo- pyrimidine-4-carboxylate (2c) roformate in presence of mild base sodium carbonate The product obtained was white solid from ethyl 2-(1- followed by the oxime formation by using hydroxyla- aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimidine- mine hydrochloride in presence of base potassium hydro- 4-carboxylate 1 (0.1 g, 0.28 mmol), 2-fluorobenzoyl xide and methanol as a solvent to get benzyl 1-amidino- chloride (0.044 g, 0.28 mmol) and triethylamine (0.085 g, cyclobutylcarbamate. This compound on cyclisation with 0.84 mmol). diethyl acetylene dicarboxylate in chloroform using trie- 2.1.6 Synthesis of Ethyl 2-(1-(4-tert-butylbenzamido) thylamine as a base for 5 hr yielded benzyl 1-(4-(ethoxy- Cyclobutyl)-5-(benzoyloxy)-6-hydroxy- carbonyl)-5, 6-dihy-droxypyrimidin-2-yl) cyclobutylcar- pyrimidine-4-carboxylate (2d) bamate. Treating the above compound with benzoic an- The product obtained was white solid from ethyl 2-(1- hydride in pyridine gave benzyl 1-(4-(ethoxycarbonyl)- aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimidine- 5-(benzoyloxy)-6-hydroxylpyrimidin-2-yl)cyclobutyl car- 4-carboxylate 1 (0.1 g, 0.28 mmol), 4-tert-butyl-benzoyl bamate. Final key intermediate 1 was obtained by depro- chloride (0.055 g, 0.28 mmol) and triethylamine (0.085 g, tection of amine group of previous compound by using 0.84 mmol).

OH Table 1. IR, 1H NMR and mass spectroscopic data of com- O O (i) N R C NH pounds 2(a-i) O N COOC2H5 OH Com- IR MS O 2(a-i) N 1 pound (KBr, H NMR (DMSO-d6, 400 MHz)  (M+1) O -1 H2N N COOC2H5 No cm ) m/z 1 OH 3460, 12.85 (s, 1H, -OH), 9.85 (s, 1H, O O (ii) N R S NH 2930, -NH), 9.65 (s, 1H, Ar-H), 9.15 (m, O O N COOC2H5 2859, 2H, Ar-H), 7.22 (m, 5H, Ar-H), 2a 552.41 2371, 4.24 (q, 2H, -CH2), 2.82 (t, 2H, 3(a-i) 1681, -CH2), 2.45 (t, 2H, -CH2), 1.97 (m, Scheme 1. Reagents and conditions: (i) Alkyl/Aryl carbonyl 1059 2H -CH2), 1.10 (t, 3H, -CH3) o chloride, triethylamine, dichloromethane, 0 C-rt, 6-8 hr; (ii) 13.08 (s, 1H, -OH), 9.01 (s, 1H, 3459, Alkyl/Aryl sulphonyl chloride, triethylamine, dichloro- -NH), 8.08 (t, 2H, Ar-H), 7.78 (d, 2941, o 1H, Ar-H), 7.63 (m, 1H, Ar-H), methane, 0 C-rt, 6-8 hr 2848, 2b 7.12 (m, 5H, Ar-H), 4.24 (q, 2H, 492.17 2365, -CH ), 3.84 (s, 3H, -OCH ), 2.82 (t, 1679, 2 3 2.1.7 Synthesis of Ethyl 2-(1-(2,6-difluorobenzamido) 2H, -CH ), 2.45 (t, 2H, -CH ), 1.97 1065 2 2 Cyclobutyl)-5-(benzoyloxy)-6-hydroxy- (m, 2H -CH2), 1.10 (t, 3H, -CH3) pyrimidine-4-carboxylate (2e) 3450, 12.89 (s, 1H, -OH), 9.80 (s, 1H, The product obtained was white solid from ethyl 2-(1- 2959, -NH), 7.78 (t, 1H, Ar-H), 7.38 (t, aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimidine- 2855, 1H, Ar-H), 7.85 (d, 1H, Ar-H), 2c 2360, 7.58 (d, 1H, Ar-H), 7.17 (m, 5H, 480.46 4-carboxylate 1 (0.1 g, 0.28 mmol), 2, 6-difluoro- 1665, Ar-H), 4.22 (q, 2H, -CH2), 2.75 (t, benzoyl chloride (0.049 g, 0.28 mmol) and triethylamine 1079, 2H, -CH2), 2.42 (t, 2H -CH2), 1.91 (0.085 g, 0.84 mmol). 790 (m, 2H, -CH2), 1.12 (t, 3H, -CH3) 12.56 (s, 1H, -OH), 10.25 (s, 1H, 2.1.8 Synthesis of Ethyl 2-(1-(3-bromobenzamido) 3462, -NH), 8.88 (d, 2H, Ar-H), 7.87 (d, 2955, Cyclobutyl)-5-(benzoyloxy)-6-hydroxy- 2H, Ar-H), 7.44-7.50 (m, 5H, 2848, pyrimidine-4-carboxylate (2f) 2d Ar-H), 4.22 (q, 2H, -CH2), 2.79 (t, 518.22 2358, 2H, -CH ), 2.50 (t, 2H, -CH ), 1.96 The product obtained was white solid from ethyl 2-(1- 1669, 2 2 (m, 2H, -CH ), 1.29 (s, 9H, aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimidine- 1045 2 -(CH ) ), 1.07 (t, 3H, -CH ) 4-carboxylate 1 (0.1 g, 0.28 mmol), 3-bromobenzoyl 3 3 3 chloride (0.061 g, 0.28 mmol) and triethylamine (0.085 g, 3434, 12.52 (s, 1H, -OH), 9.85 (s, 1H, 2927, -NH), 7.91 (dd, 2H, Ar-H), 7.75 0.84 mmol). 2360, (m, 1H, Ar-H), 7.23 (m, 5H, Ar-H), 2e 498.14 2.1.9 Synthesis of Ethyl 2-(1-(4-chlorobenzamido) 1666, 4.25 (q, 2H, -CH2), 2.76 (t, 2H, 1105, -CH ), 2.53 (t, 2H, -CH ), 1.94 (m, Cyclobutyl)-5-(benzoyloxy)-6-hydroxy- 2 2 794 2H, -CH2), 1.1 (t, 3H -CH3) pyrimidine-4-carboxylate (2g) 3451, 12.95 (s, 1H, -OH), 9.20 (s, 1H, The product obtained was white solid from ethyl 2-(1- 2956, -NH), 8.24 (s, 1H, Ar-H), 8.06 (d, aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimidine- 2854, 1H, Ar-H), 7.88 (t, 1H, Ar-H), 7.61 4-carboxylate 1 (0.1 g, 0.28 mmol), 4-chlorobenzoyl 2f 2362, (m, 5H, Ar-H), 7.48 (d, 1H, Ar-H), 541.07 chloride (0.049 g, 0.28 mmol) and triethylamine (0.085 g, 1672, 4.21 (q, 2H, -CH2), 2.78 (t, 2H, 0.84 mmol). 1046, -CH2), 2.48 (t, 2H -CH2), 1.95 (m, 763 2H, -CH2), 1.01 (t, 3H, -CH3) 2.1.10 Synthesis of Ethyl 2-(1-(benzamido) 3458, 12.43 (s, 1H, -OH), 10.21 (s, 1H, Cyclobutyl)-5-(benzoyloxy)-6-hydroxy- 2951, -NH), 8.91 (d, 2H, Ar-H), 7.93 (d, 2836, 2H, Ar-H), 7.44-7.56 (m, 5H, pyrimidine-4-carboxylate (2h) 2g 496.89 The product obtained was white solid from ethyl 2-(1- 2355, Ar-H), 4.29 (q, 2H, -CH2), 2.73 (t, 1668, 2H, -CH ), 2.48 (t, 2H, -CH ), 1.94 aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimidine- 2 2 723 (m, 2H, -CH2), 1.05 (t, 3H, -CH3) 4-carboxylate 1 (0.1 g, 0.28 mmol), benzoyl chloride 3450, 12.79 (s, 1H, -OH), 9.31 (s, 1H, (0.039 g, 0.28 mmol) and triethylamine (0.085 g, 0.84 2960, -NH), 7.52-7.85 (m, 10H, Ar-H), mmol). 2h 2849, 4.25 (q, 2H, -CH2), 2.80 (t, 2H, 462.26 2.1.11 Synthesis of Ethyl 2-(1-(3-nitrobenzamido) 2356, -CH2), 2.49 (t, 2H, -CH2), 2.02 (m, 1679 2H, -CH ), 1.06 (t, 3H, -CH ) Cyclobutyl)-5-(benzoyloxy)-6-hydroxy- 2 3 12.70 (s, 1H, -OH), 9.25 (s, 1H, 3462, pyrimidine-4-carboxylate (2i) -NH), 8.22 (s, 1H, Ar-H), 8.12 (d, 2941, The product obtained was white solid from ethyl 2-(1- 1H, Ar-H), 7.91 (t, 1H, Ar-H), 7.63 2863, aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimi-dine 2i (m, 5H, Ar-H), 7.43 (d, 1H, Ar-H), 507.34 2358, 4.24 (q, 2H, -CH ), 2.81 (t, 2H, -4-carboxylate 1 (0.1 g, 0.28 mmol), 3-nitrobenzoyl chlo- 1685, 2 -CH ), 2.50 (t, 2H -CH ), 1.98 (m, 1061 2 2 ride (0.052 g, 0.28 mmol) and triethylamine (0.085 g, 2H, -CH ), 1.12 (m, 3H, -CH ) 0.84 mmol). 2 3

2.2 Biology causing 50% inhibition of cell growth) were estimated after 72 h of compound treatment. The absorbance of Human cell lines, K562 and CEM were purchased from vehicle cells was taken as 100% viability and the values National Center for Cell Science, Pune, India. Cells were of treated cells were calculated as a percentage of control cultured in RPMI 1640 medium supplemented with 10% and presented as histograms (Figure 1 and Figure 2). fetal bovine serum (FBS), 100 U/mL of Penicillin, and The antiproliferative effects of these compounds were 100 μg of streptomycin/ml and incubated at 37 ºC in a also analyzed by trypan blue exclusion assay against both humidified atmosphere containing 5% CO2. Leukemia K562 and CEM cells (data not shown). (K562 and CEM) cells growing in log phase were treated with 10, 50 and 100 μM concentrations of 2, 4, 5, 2.2.2 Cell Cycle Analysis 6-substituted pyrimidine derivatives 2(a-i) and 3(a-i). Cellular DNA content was measured by flow cytometry. Since the compounds were dissolved in DMSO, it was Approximately 7.5×104 cells/ml were cultured and used as vehicle control. We employed trypan blue dye treated with 10, 50 and 100 μM concentrations of 2e or exclusion and MTT assay to assess the cytotoxicity. In 3f (Figure 3). Cells were harvested after 48 h of treat- addition, we have performed cell cycle analysis and ment, washed, fixed in 70% ethanol and incubated with DNA fragmentation assay to determine the apoptosis. RNase A (Sigma-Aldrich, USA). Propidium iodide (PI, Each experiment was repeated a minimum of two times. 50 μg/ml, Sigma-Aldrich, USA) was added half an hour before acquiring the flow cytometric reading (FACScan, 2.2.1 Cytotoxicity Assay BD Biosciences, USA). A minimum of 10,000 cells were MTT [3-(4, 5 dimethylthiazol-2-yl-2,5-tetrazolium bro- acquired per sample and histograms were analyzed by mide)] assay was used for the measurement of the cyto- using WinMDI 2.8 software. toxicity of synthesized compounds 2(a-i) and 3(a-i) as described previously [19]. In brief, Exponentially grow- 3. Results and Discussion ing K562 or CEM cells (1×104 cells /well) were plated in duplicates and incubated with 10, 50 and 100 μM of 2(a-i) Pyrimidine ring was derivatized by substituting electron and 3(a-i). Cells were harvested after 48 and 72 h of withdrawing and electron releasing groups along with treatment and incubated with MTT (0.5 mg/ml) at 37°C. cyclobutyl carboxamide 2(a-i) and sulfonamide 3(a-i) The water soluble tetrazolium salt, [3-(4, 5-dimethylthi- moiety at position 2 of the pyrimidine ring. In addition, azol-2-yl-2,5-tetrazolium bromide)] is metabolized to the we also introduced hydroxyl group at position 6, ethyl water insoluble formazan by intact mitochondrial dehy- carboxylate at position 4 and phenyl carboxylate at posi- drogenases. The formazan is then solubilized by adding tion 5 of the pyrimidine. To investigate the cytotoxic detergent. The viability of the cells was estimated on the effects of 2,4,5,6-tetrasubstituted pyrimidines on the basis of formazan formed, which was detected spectro- growth of leukemia cells, a dose response study was photometrically by optical density at 570 nm. The mean conducted using trypan blue dye exclusion (data not absorbance of culture medium was used as the blank and shown) and MTT assay (Figure 1 and Figure 2). Results was subtracted. IC50 values (concentration of compound showed that, the cytotoxicity induced by the derivatives

(a) (b) Figure 1. Cytotoxicity as measured by MTT assay. K562 cells treated with 10, 50 and 100 µM of compounds 2(a-i) (a) and 3(a-i) (b) for 48 and 72 h were harvested and used for the assay. DMSO treated cells were used as vehicle control. Data are representative of the mean of two separate experiments done in duplicate

(a) (b) Figure 2. Cytotoxicity as measured by MTT assay. CEM cells treated with 10, 50 and 100 µM of compounds 2(a-i) (a) and 3(a-i) (b) for 48 and 72 h were harvested and used for the assay. DMSO treated cells were used as vehicle control. Data are representative of the mean of two separate experiments done in duplicate

(a) (b)

(c) (d) Figure 3. Cell cycle analysis of K562 cells treated with 2e or 3f. K562 cells (0.75×105 cells/ml) were incubated at 37 °C with 2e or 3f (10, 50 and 100 μM). Following 48 h of incubation, cells were fixed and stained with propidium iodide and subjected to FACS analysis. Panel (a) and (b) show histograms comparing the effect of 2e and 3f at specific cell cycle stages. In both the Panels (a) and (b), the first histogram represents DMSO treated cells. Panel (c) and (d) show the quantification of cells in different stages of cell cycle followed by treatment with 2e and 3f respectively

2(a-i) and 3(a-i) was in a concentration and time depend- taken as 100%. The relative percentage was calculated ent manner. Interestingly, the DMSO control corre- for the treated compounds (Figure 1 and Figure 2). sponding to the highest concentrations of compounds These results were further confirmed by trypan blue as- tested did not show any significant toxic effect and it was say (data not shown). Based on these results IC50 values

were calculated for 72 h and tabulated in Table 2. Com- Table 2. Structure and IC50 values of the synthesised compounds 2c, 2d, 2e, 2f, 2g and 2i in carboxamide series pounds 2(a-i) and 3(a-i) and compounds 3b, 3c, 3d, 3f and 3i in sulphonamide IC50 (μM) series showed good cytotoxicity. As can be seen from Compound R Table 2, the electronic property and position of the group K562 CEM on the phenyl ring attached to the carboxamide/sulfonamide O2N moiety determines the activity of these compounds. 2a 45.2± 6.8 54.5± 7.2 Among carboxamide derivatives 2(a-i), we found that NO compounds with halogen, nitro and tert-butyl as sub- 2 stituents on phenyl ring are more cytotoxic compared to 2b >100 >100 unsubstituted (2h) and methoxy substituted (2b) deriva- OMe tives. Compound 2g with chloro at para position on the 2c 27.6± 2.6 32.2± 3.8 phenyl ring of carboxamide showed good activity with an F F IC50 of 14.0 µM against K562 cells. Among halogen substitution containing sulfonamide derivatives, compound 2d 36.1± 3.4 24.3± 2.3 3c with chloro groups at two meta positions showed good F activity (IC50: 15.0 µM) compared to a chloro at para H C position (3d, IC50: 28 µM) for K562 cells. Similarly, in 2e 3 40.0± 5.2 34.7± 3.0 nitro group containing sulfonamide derivatives activity H3C CH3 varies with the position. Compound 3b with nitro group at meta position exhibited more activity (IC50: 26.2 µM) 2f 50.2± 7.1 22.1± 2.1 relative to compound 3e with an ortho nitro group (IC50: Br 36.5 µM) and compound 3a with a para nitro (IC50: 43.1 µM) against K562 cells. Besides electron withdrawing, electron releasing groups also played a role in enhancing 2g 14.0± 2.1 23.0± 2.4 Cl the activity. Compound 3f with tert-butyl group at para position 2h >100 >100 exhibited more activity than 3i containing methyl group at the same position. This could be due to the presence of three electron releasing groups (methyl groups) in 3f and 2i 35.5± 3.6 48.6± 7.2 only one electron releasing group (methyl group) in 3i. NO More importantly, alkyl group (methyl) directly attached 2 to the sulfur atom of the sulfonamide also showed modest 3a 43.1± 6.5 52.3± 7.8 activity. Interestingly, compound 3g containing only O2N phenyl ring lost the activity. This suggest that, electron O2N releasing or electron withdrawing group on the phenyl 3b 26.2± 3.2 45.1± 6.8 ring, or the groups directly attached to the sulfonamide Cl moiety contributed to the enhancement of cytotoxicity of 3c 15.0± 1.6 25.3± 3.8 these compounds. Our previous study on the cytotoxic effect of thia- Cl zolidinone derivatives suggest that, the electron donating 3d 28.4± 4.1 30.3± 4.3 th groups on the C-terminal of the phenyl group at 4 posi- Cl tion resulted in increasing the activity [32]. Inspired by these results, in this series, we have chosen two mole- 3e 36.5± 4.9 42.1± 6.5 NO2 cules 2e and 3f containing tert-butyl group at 4th position H C from carboxamide and sulfonamide derivatives respec- 3f 3 20.0± 2.3 24.0± 3.0 tively for further studies. Firstly, to evaluate the effect of H3C CH3 2e and 3f on cell cycle progression we have carried out flow cytometry analysis. Results showed that both 2e and 3g >100 >100 3f did not induce any apoptosis up to 50 µM. At 100 µM we have seen a remarkable accumulation of subG1 cells 3h H3C 30.2± 4.3 32.3± 4.4 followed by the decline of G1, S and G2/M phase cells (Figure 3). Therefore, our studies further confirm that 3i 24.1± 3.1 20.0± 3.4 growth inhibition could be due to apoptosis. H3C

4. Conclusions substituted 4-anilino-2-(2-pyridyl) pyrimidines as a new series of apoptosis inducers using a cell- and caspase-based In summary, a series of 2, 4, 5, 6-substituted pyrimidine high throughput screening assay,” Part 1: Structure-activity derivatives were synthesized and evaluated for antiprolif- relationships of the 4-anilino group, Bioorganic and Me- erative activity against human leukemia cells. From the dicinal Chemistry Letters, Vol. 14, pp. 7761–7773, 2006. current investigation, structure–activity relationships of [6] S. S. Bahekar and D. B. Shinde, “Synthesis and anti-in- those compounds suggest both electron donating and elec- flammatory activity of some [4,6-(4-substituted aryl)-2- tron withdrawing groups on the phenyl ring attached to the thioxo-1,2,3,4-tetrahydro-pyrimidin-5-yl]-acetic acid deri sulfonamide group will determine the anticancer activity. vatives,” Bioorganic and Medicinal Chemistry Letters, Compounds with halogen substitution at different positions Vol. 14, pp. 1733–1736, 2004. on the phenyl ring of the aryl carboxamide and sulphona- [7] K. Tadashi, I. Motonobu, H. Akio, K. Fumihiko, and K. mide showed good cytotoxicity. From cell cycle analysis, Kazuo, “Pyrimidine nucleosides. 6. Synthesis and anti- it is confirmed that tert-butyl group containing derivatives cancer activities of N4-substituted 2,2'-anhydronucle- able to could induce apoptosis. Further detailed investiga- osides,” Journal of Medicinal Chemistry, Vol. 17, pp. tion on the structure- activity relationship should consider 1076–1078, 1974. the substitution pattern on phenyl ring as a means to lead [8] S. L. Tai, H. Y. Jing, C. L. Mao, Y. S. Zh, C. C. Yung, H. the discovery for more potent cytotoxic compounds. Stud- P. William, I. B. George, G. Jerzy, and G. Ismail, “Synth- ies on the mechanism of action these compounds and esis and anticancer activity of various 3'-deoxy pyrimidine modification is under progress. nucleoside analogs, and crystal structure of 1-(3-deoxy-. beta.-D-threo-pentofuranosyl) cytosine,” Journal of Medi- 5. Acknowledgements cinal Chemistry, Vol. 34, pp. 693–701, 1991. [9] N. Ebrahim, Z. Aihua, K. Panteh, I. W. Leonard, B. Jan, The authors are grateful to UGC, Govt. of India for finan- D. C. Erik, and E. K. Edward, “Synthesis of 3‘- and cial support to KSR under the UGC-SAP (DRS-II) vide 5‘-nitrooxy pyrimidine nucleoside nitrate esters: ‘Nitric No. F. 540/10/2004-05 Programme. Asha D is grateful Oxide Donor’ agents for evaluation as anticancer and an- to SC/ST Cell, University of Mysore for JRF and UGC, tiviral agents,” Journal of Medicinal Chemistry, Vol. 46, New Delhi for fellowship under RFSMS order No. pp. 995–1004, 2003. DV5/373[3]/RFSMS/2008-09. Sathees C Raghavan ac- [10] G. N. Pershin, L. I. Sherbakova, T. N. Zykova, and V. N. knowledges the support from Lady Tata Memorial Trust Sakolova, “Antibacterial activity of pyrimidineand pyr- international award for leukemia research (London). rolo-(3, 2-d)-pyrimidine derivatives,” Farmakol Taksikol, Prasanna DS is grateful to Council of Scientific and In- Vol. 35, pp. 466–471, 1972. dustrial Research, New Delhi for financial support under [11] D. J. Brown, “Pyrimidines and their benzo derivatives,” CSIR-SRF order No. 09/119(0173)2K8 EMR-I. In: A. R. Katrizky, C. W. Rees Ed., Comprehensive heterocyclic chemistry, the structure, reactions, synthesis and REFERENCES uses of heterocyclic compounds, Pragmon Press, Oxford, pp. 57, 1984. [1] J. B. Gibbs, “Mechanism-based target identification and [12] N. I. Smetskaya, N. A. Mukhina, V. G. Granik, G. Y. drug discovery in cancer research,” Science, Vol. 287, pp. Shvarts, R. D. Syubaev, and M. D.Mashkovskii, “Synthe- 1969–1973, 2000. sis and study of analgesic action of pyrido [1,2-a] pyri- [2] B. A. Chabner, W. Wilson, and J. Supko, “Pharmacology midine derivatives,” Pharmaceutical Chemistry Journal, and toxicity of antineoplastic drugs,” In: E. Beutler, M. A. Vol. 18, pp. 540–544, 1984. Lichtman, B. S. Coller, T. J. Kipps, U. Seligsoh, Ed., Wil- [13] S. Naresh, Nishi, P. Shraddha, M. S. C. Prem, and G. liams Hematology, 6th edition, McGraw Hill, New York, Suman, “Synthesis and antileishmanial activity of novel pp. 185, 2001. 2,4,6-trisubstituted pyrimidines and 1,3,5-triazines,” Euro- [3] X. Fuchun, Z. Hongbing, Z. Lizhi, L. Liguang, and H. pean Journal of Medicinal Chemistry, Vol. 44, pp. 2473– Youhong, “Synthesis and biological evaluation of novel 2481, 2009. 2,4,5-substituted pyrimidine derivatives for anticancer activity,” Bioorganic and Medicinal Chemistry Letters, Vol. [14] A. Z. Medhat, M. S. Ahmed, S. A. E. Mohamed, A. A. 19, pp. 275–278, 2009. Yousry, and H. E. Usama, “Some reactions with Ketene dithioacetals: Part I: Synthesis of antimicrobial pyrazolo [1,5-a] [4] D. G. Monica, A. G. V. Jose, R. S. Fernando, A. M. Juan, pyrimidines via the reaction of ketene dithioacetals and C. Octavio, A. Antonia, A. G. Miguel, E. Antonio, and M. 5-aminopyrazoles,” Il Farmaco, Vol. 56, pp. 277–283, 2001. C. Joaquin, “Anticancer activity of (1,2,3,5-tetrahydro-4, 1-benzoxazepine-3-yl)-pyrimidines and -purines against [15] E. T. Buurman, A. E. Blodgett, K. G. Hull, and D. Carca- the MCF-7 cell line: Preliminary cDNA microarray stud- nague, “Pyridines and pyrimidines mediating activity ies,” Bioorganic and Medicinal Chemistry Letters, Vol. against an efflux-negative strain of candida albicans 18, pp. 1457–1460, 2008. through putative inhibition of lanosterol demethylase,” An- timicrob Agents Chemother, Vol. 48, pp. 313–318, 2004. [5] S. Nilantha, K. Shailaja, N. Bao, P. Azra, W. Yang, C. Gisela, T. Ben, D. John, and X. C. Sui, “Discovery of [16] P. J. Manley, A. E. Balitza, M. T. Bilodeau, K. E. Coll, G.

D. Hartman, R. C. McFall, K. W. Rickert, L. D. Rodman, EGFR selective 4,6-disubstituted pyrimidines from a and K. A. Thomas, “2, 4-Disubstituted pyrimidines: A combinatorial kinase-directed heterocycle library,” Jour- novel class of KDR kinase inhibitors,” Bioorganic and Me- nal of the American Chemical Society, Vol. 128, pp. dicinal Chemistry Letters, Vol. 13, pp. 1673–1677, 2003. 2182–2183, 2006. [17] M. Anderson, J. F. Beattie, G. A. Breault, J. Breed, K. F. [25] V. J. Cee, B. K. Albrecht, S. Geuns-Meyer, P. Hughes, S. Byth, J. D. Culshaw, R. P. A. Ellston, S. Green, C. A. Bellon, J. Bready, S. Caenepeel, S. C. Chaffee, A. Coxon, Minshull, R. A. Norman, R. A. Pauptit, J. Stanway, A. P. M. Emery, J. Fretland, P. Gallant, Y. Gu, B. L. Hodous, D. Thomas, and P. J. Jewsbury, “Imidazo [1,2-a] pyridines: A Hoffman, R. E. Johnson, R. Kendall, J. L. Kim, A. M. potent and selective class of cyclin-Dependent kinase inhibi- Long, D. McGowan, M. Morrison P. R. Olivieri, V. F. Patel, tors identified through structure-Based hybridization,” A. Polverino, D. Powers, P. Rose, L. Wang, and H. Zhao, Bioorganic and Medicinal Chemistry Letters, Vol. 13, pp. “Alkynylpyrimidine amide derivatives as potent, selective, 3021–3026, 2003. and orally active inhibitors of tie-2 kinase,” Journal of Me- [18] K. F. Byth, J. D. Culshaw, S. Green, S. E. Oakes, and A. P. dicinal Chemistry, Vol. 50, pp. 627–640, 2007. Thomas, “Imidazo [1,2-a] pyridines. Part 2: SAR and opti- [26] G. W. Rewcastle, W. A. Denny, and H. D. H. Showalter, misation of a potent and selective class of cyclin-dependent “Synthesis of 4-(Phenylamino) pyrimidine derivatives as kinase inhibitors,” Bioorganic and Medicinal Chemistry atp-competitive protein kinase inhibitors with potential Letters, Vol. 14, pp. 2245–2248, 2004. for cancer chemotherapy,” Current Organic Chemistry, [19] S. Emanuel, R. H. Gruninger, A. Fuentes-Pesquera, P. J. Vol. 4, pp. 679–706, 2000. Connolly, J. A. Seamon, S. Hazel, R. Tominovich, B. [27] A. J. Bridges, “Chemical inhibitors of protein kinases,” Hollister, C. Napier, M. R. D’Andrea, M. Reuman, G. Chemical Reviews, Vol. 101, pp. 2541–2572, 2001. Bignan, R. Tuman, D. Johnson, D. Moffatt, M. Batchelor, A. Foley, J. O’Connell, R. Allen, M. Perry, L. Jolliffe, [28] P. Traxler, E. Bold, G. Buchdunger, G. Caravatti, P. Furet, and S. A. Middleton, “A vascular endothelial growth P. Manley, T. O’Reilly, J. Wood, and J. Zimmermann, factor receptor-2 kinase inhibitor potentiates the activity “Tyrosine kinase inhibitors: From rational design to of the conventional chemotherapeutic agents paclitaxel clinical trials,” Medicinal Research Reviews, Vol. 21, pp. and doxorubicin in tumor xenograft models,” Molecular 499–512, 2001. Pharmacology, Vol. 66, pp. 635, 2004. [29] D. S. Prasanna, C. V. Kavitha, B. Raghava, K. Vinaya, S. [20] J. T. Sisko, T. J. Tucker, M. T. Bilodeau, C. A. Buser, P. A. R. Ranganatha, S. C. Raghavan, and K. S. Rangappa, Ciecko, K. E. Coll, C. Fernandes, J. B. Gibbs, T. J. Koester, “Synthesis and identification of a new class of (S)-2,6- N. Kohl, J. J. Lynch, X. Mao, D. McLoughlin, C. M. diamino-4,5,6,7-tetrahydrobenzo[d]thiazole derivatives as Miller-Stein, L. D. Rodman, K. W. Rickert, L. potent antileukemic agents,” Investigational New Drugs, Sepp-Lorenzino, J. M. Shipman, K. A. Thomas, B. K. 2009. Wong, and G. D. Hartman, “Potent 2-[(pyrimidin-4-yl) [30] C. S. Ananda-Kumar, C. V. Kavitha, K. Vinaya, S. B. amine}-1,3-thiazole-5-carbonitrile-based inhibitors of Benaka Prasad, N. R. Thimmegowda, S. Chandrappa, S. VEGFR-2 (KDR) kinase,” Bioorganic and Medicinal C. Raghavan, and K. S. Rangappa, “Synthesis and in vitro Chemistry Letters, Vol. 16, pp. 1146–1150, 2004. cytotoxic evaluation of novel diazaspiro bicyclo hydan- [21] A. G. Waterson, K. L. Stevens, M. J. Reno, Y. M. Zhang, E. toin derivatives in human leukemia cells: A SAR study,” E. Boros, F. Bouvier, A. Rastagar, D. E. Uehling, S. H. Investigational New Drugs, Vol. 27, pp. 327–337, 2009. Dickerson, B. Reep, O. B. McDonald, E. R. Wood, D. W. [31] C. V. Kavitha, M. Nambiar, C. S. Ananda-Kumar, B. Rusnak, K. J. Alligood, and S. K. Rudolph, “Alkynyl Choudhary, K. Muniyappa, K. S. Rangappa, and S. C. pyrimidines as dual EGFR/ErbB2 kinase inhibitors,” Bio- Raghavan, “Novel derivatives of spirohydantoin induce organic and Medicinal Chemistry Letters, Vol. 16, pp. growth inhibition followed by apoptosis in leukemia cells,” 2419–2422, 2006. Biochemical Pharmacology, Vol. 77, pp. 348–363, 2009. [22] G. A. Breault, R. P. A. Ellston, S. Green, S. R. James, P. J. [32] S. Chandrappa, C. V. Kavitha, M. S. Shahabuddin, K. Jewsbury, C. J. Midgley, R. A. Pauptit, C. A. Minshull, J. Vinaya, C. S. Ananda-Kumar, S. R. Ranganatha, S. C. A. Tucker, and J. E. Pease, “Cyclin-dependent kinase 4 Raghavan, and K. S. Rangappa, “Synthesis of 2-(5-((5-(4- inhibitors as a treatment for cancer. Part 2: Identification chlorophenyl)furan-2-yl) methylene)-4-oxo-2- thioxothi- and optimisation of substituted 2, 4-bis anilino pyrimi- azolidin-3-yl)acetic acid derivatives and evaluation of dines,” Bioorganic and Medicinal Chemistry Letters, Vol. their cytotoxicity and induction of apoptosis in human 13, pp. 2961–2966, 2003. leukemia cells,” Bioorganic and Medicinal Chemistry, [23] J. F. Beattie, G. A. Breault, R. P. A. Ellston, S. Green, P. J. Vol. 17, pp. 2576–2584, 2009. Jewsbury, C. J. Midgley, R. T. Naven, C. A. Minshull, R. [33] D. Asha, Manish Malviya., S. Chandrappa, , C. T. Sada- A. Pauptit, J. A. Tucker, and J. E. Pease, “Cyclin-depen- shiva, K. Vinaya, D. S. Prasanna and K. S. Rangappa, dent kinase 4 inhibitors as a treatment for cancer. Part 1: “Synthesis and characterization of substituted ethyl 2- Identification and optimisation of substituted 4, 6-Bis ani- (1-aminocyclobutyl)-5-(benzoyloxy)-6-hydroxypyrimi-di lino pyrimidines,” Bioorganic and Medicinal Chemistry ne-4-carboxylate derivatives as antioxidant agents,” Let- Letters, Vol. 13, pp. 2955–2960, 2003. ters in Drug Design and Discovery, Vol. 6, No. 8, pp. [24] Q. Zhang, Y. Liu, F. Gao, Q. Ding, C. Cho, W. Hur, Y. 637–643, 2009. Jin, T. Uno, C. A. P. Joazeiro, and N. Gray, “Discovery of

Transfusion of Ipscs-Derived Leukocytes to Kill Cancer

Jiang Li1*, Ying Cui2, Guodong Gao1*, Ti-Fei Yuan

1Department of Neurosurgery, Tang Du Hospital, The Fourth Military University, Xi’an, China; 2Department of Clinical Laboratory and Research center of Tang Du Hospital, The Forth Military Medical University; Xi’an, China; *Corresponding author Email: lijiang19771103@gmail, [email protected], [email protected]

Received October 28th, 2009; revised December 23rd, 2009; accepted January 23rd, 2010.

ABSTRACT Here we propose that the rejuvenation of leukocytes with iPSC technology in vitro and transfusion of cancer cell- resistant white blood cells back to human body provide a prospective therapy for cancer patients.

Keywords: Ipscs, Leukocytes, Cancer

1. Introduction 2. IPSC Technology 60 years ago, Chester Southam injected healthy people The regenerative medicine or rejuvenation therapy re- and cancer patients with Hela cells, and found that quires a large number of health cells to be placed in healthy human beings are resistant to cancer cells, the damaged parts of the body. However only limited sources ability of which was lost in cancer patients [1]. The im- of adult stem cells are available, and the self-repair abil- portant factors that lead to the loss of innate immunity ity in mammals including human is very restricted. The during cancer progression include the failure of leuko- use of embryonic stem cells suffers from ethical difficul- cytes (white blood cells) in killing cancer cells and the ties, and the transplantation of heterogeneous tissues of- fusion of cancer cell and leukocyte, which results in the ten meets immunological challenges. In the last two years, metastasis of cancer cells [2]. In recent years, a sponta- the successes of inducible pluripotent stem cell (iPSC) neous cancer-resistant mice line was found, whose resis- technology on human cells [4,5], including patients- tance is mediated by rapid infiltration of leukocytes and sourced cells [6], provide new hopes in generating im- some other cells with innate immunity [3]. In these mice, mune-free stem cell lines in treating human patients. the killing of injected cancer cells was completed by the iPSC can be acquired by transfection of somatic cell migration of leukocytes to the cancer cells-injected site, with pluripotent factors-coding viral vectors, or the direct the identification/recognition of specific antigens on the delivery of DNA plasmids into cytoplasm to promote a surface of cancer cells, as well as the release of cancer pluripotent fate [4,5]. Recent studies found that the tran- cell-killing substances, the first two steps of which do not sient delivery of these proteins without genome integra- exist in normal mice lines [3]. Additionally, aging leads tion can also bring pluripotency [7,8]. Additionally, nu- to the reduction of this resistance ability, while repeated merous small molecules have been identified to promote challenge with cancer cell injection can enhance the re- the efficiency in generating iPSCs [9,10], including natu- sistance during ageing periods. It is thus conceivable that ral compounds such as Vitamin C [11]. Methods with the failed or weakened resistance of innate immunity further improved safety and efficiency are expected in cells, especially leukocytes, contribute to the cancer pro- coming years. gress in patients. The iPSCs, once stabilized, can be expanded greatly To restore the innate immunity, efforts must be made with many kinds of trophic factors and cytokines. More- to re-activate the activity of white blood cells in pa- over, additional genetic engineering could be performed tients with cancer, or to replace those “sick” leukocytes to repair potential genetic defects of the patients, or to with healthy, young leukocytes engineered in vitro, the promote the differentiation of iPSC into a defined fate. It process of which can now be completed with the re- will be therefore possible to acquire large number of cells, cently emerged iPSC technology. such as leukocytes, for repeated clinical use.

3. The Hypothesis Alexander-Miller, C. Von Kap-Herr, M. J. Pettenati, A. M. Sanders, H. M. Weir, W. Du, J. Kim, et al., “Transferable We hypothesize that the transfusion of healthy, cancer anticancer innate immunity in spontaneous regres- cell-resistant leukocytes (in future, may include other sion/complete resistance mice,” Proceedings of the Na- innate immunity associated cells) generated with iPSC tional Academy of Science, USA, Vol. 103, No. 20, pp. technology can be used to treat cancer patients. More 7753–7758, 2006. importantly, genetic modified leukocytes (such as the [4] J. Yu, M. A. Vodyanik, K. Smuga-Otto, J. Antosiewicz- knockout of specific receptors [12]) could be resistant to Bourget, J. L. Frane, S. Tian, J. Nie, G. A. Jonsdottir, V. cancer cell fusion, therefore stopping the progression of Ruotti, R. Stewart, et al., “Induced pluripotent stem cell cancer cells in vivo. lines derived from human somatic cells,” Science, Vol. However it should be also noted that white blood cells 318, No. 5858, pp. 1917–1920, 2007. family contains several types of cells such as neutrophils, [5] K. Takahashi, K. Tanabe, M. Ohnuki, M. Narita, T. Ichi- eosinophils and basophils; these cells are of diverse func- saka, K. Tomoda, and S. Yamanaka, “Induction of pluri- tion within the tumor microenvironment. For example, a potent stem cells from adult human fibroblasts by defined recent study identified different subpopulations of neu- factors,” Cell, Vol. 131, No. 5, pp. 861–872, 2007. trophils (N1-N2) that may inhibit or promote the tumor [6] G. Amabile and A. Meissner, “Induced pluripotent stem growth [13]. Therefore it will be useful to genetically cells: Current progress and potential for regenerative engineer only a subpopulation of one kind of leukocytes medicine,” Trends in Molecular Medicine, Vol. 15, No. 2, in reaching effective tumor control. pp. 59–68, 2009. [7] D. Kim, C. H. Kim, J. I. Moon, Y. G. Chung, M. Y. 4. Testing the Hypothesis and Clinical Chang, B. S. Han, S. Ko, E. Yang, K. Y. Cha, R. Lanza, Significance et al., “Generation of human induced pluripotent stem cells by direct delivery of reprogramming proteins,” Cell There’re few technical difficulties to generate iPSCs from Stem Cell, Vol. 4, No. 6, pp. 472–476, 2009. cancer patients and promote these cells into a leukocyte [8] H. Zhou, S. Wu, J. Y. Joo, S. Zhu, D. W. Han, T. Lin, S. cell fate. The fibroblasts from patients’ skin tissues could Trauger, G. Bien, S. Yao, Y. Zhu, et al., “Generation of be most easily harvested and induced into iPSCs with induced pluripotent stem cells using recombinant pro- well-built protocols, while theoretically all types of so- teins,” Cell Stem Cell, Vol. 4, No. 5, pp. 381–384, 2009. matic cells from biopsies can be used. Some left ques- [9] T. Lin, R. Ambasudhan, X. Yuan, W. Li, S. Hilcove, R. tions are the efficiency in expanding the precursor cells Abujarour, X. Lin, H. S. Hahm, E. Hao, A. Hayek, et al., and directed differentiation from iPSCs into leukocytes. “A chemical platform for improved induction of human These procedures could benefit from protocols inducing iPSCs,” Nature Methods, Vol. 6, No. 11, pp. 805–808, ESCs into leukocytes [14]. 2009. It will be useful to start with the mice model, to com- [10] Y. Shi, Y. ZHao, and H. Deng, “Powering reprogramming pare the efficiency of healthy leukocyte transfusion with with vitamin C,” Cell Stem Cell, Vol. 6, pp.1–2, 2010. other therapies in stopping cancer progression and me- [11] M. A. Esteban, T. Wang, B. Qin, J. Yang, D. Qin, J. Cai, tastasis. One question to be determined is to what extent W. Li, Z. Weng, J. Chen, S. Ni, et al., “Vitamin C En- the defected resistance of white blood cells contributes to hances the Generation of Mouse and Human Induced the cancer cell progress in human patients. Should one Pluripotent Stem Cells,” Cell Stem Cell, 2009. kind of cells/subpopulation of cells or cellular signaling [12] J. M. Pawelek and A. K. Chakraborty, “Fusion of tumour pathways be determined to drive tumor progression in cells with bone marrow-derived cells: A unifying expla- humans, surely the new therapy with genetically engi- nation for metastasis,” Nature Reviews Cancer, Vol. 8, No. neered iPSC-derived white blood cell transfusion would 5, pp. 377–386, 2008. follow. [13] Z. G. Fridlender, J. Sun, S. Kim, V. Kapoor, G. Cheng, L. Ling, G. S. Worthen, and S. M. Albelda, “Polarization of REFERENCES tumor-associated neutrophil phenotype by TGF-beta: “N1” versus "N2" TAN,” Cancer Cell, Vol. 16, No. 3, pp. [1] A. E. Moore, C. P. Rhoads, and C. M. Southam, “Homo- 183–194, 2009. transplantation of human cell lines,” Science, Vol. 125, No. 3239, pp. 158–160, 1957. [14] M. Hannig, H. R. Figulla, H. Sauer, and M. Wartenberg, “Control of leukocyte differentiation from embryonic [2] J. M. Pawelek and A. K. Chakraborty, “The cancer cell- stem cells upon vasculogenesis and confrontation with leukocyte fusion theory of metastasis,” Advances in Can- tumour tissue,” Journal of Cellular and Molecular Medi- cer Research, Vol. 101, pp. 397–444. 2008. cine, 2008. [3] A. M. Hicks, G. Riedlinger, M. C. Willingham, M. A.

Ovarian Sex Cord-Stromal Tumors in Postmenopausal Women and Total Laparoscopical Management

Andrea Tinelli1, Marcello Pellegrino2, Vincenzo Emanuele Chiuri3, Antonio Malvasi4

1Department of Gynecology and Obstetric, Vito Fazzi Hospital, Lecce, Italy; 2Department of Pathology, Vito Fazzi Hospital, Lecce, Italy; 3Department of Oncology, Vito Fazzi Hospital, Lecce, Italy; 4Department of Gynecology and Obstetric, Santa Maria Hospital, Bari, Italy. Email: [email protected]

Received October 26th, 2009; revised December 19th, 2009; accepted January 28th, 2010.

ABSTRACT BACKGROUND: Ovarian sex-cord stromal tumors (SCST) take up 5% of the ovarian neoplasm and may develop into an ovarian mass or a haemoperitoneum. The surgical management of SCST in early-stage adult patients is not well defined. CASE REPORT: A 69 year-old postmenopausal woman was admitted for metrorrhagia, a right ovary mass and increasing pelvic pain. Preoperative clinical and instrumental examination suspected an ovarian tumor, and the laparoscopic right ophorectomy and the frozen section suggested an ovarian SCST. To fast restore and preserve woman integrity, total laparoscopic hysterectomy (TLH) plus left salpingo-ophorectomy (SO) were performed, without compli- cations in the short and long term follow-up. CONCLUSION: In the authors’ opinion, the minimally invasive management of SCST by TLH plus bilateral SO followed by a prolonged surveillance and without intensive surgical staging, could be an appropriate clinical and surgical choice in elder patient at early stage, since these tumors are slow at growth, recurring locally and only a long time after initial treatment. We suggest, after a minimally invasive treatment, a possible “wait and see” option, as in our case report.

Keywords: Menopause, Laparoscopy, Ovarian Cancer, Sex Cord-Stromal Ovarian Tumors, Granulosa Cell Tumors, Minimally Invasive Treatment.

1. Introduction menopause, but it can also occur in puberty, associated with hyperestrogenism in 75% of total cases, leading to Ovarian sex-cord stromal tumors represent 5% of ovarian precocious puberty in children and metrorrhagia in neoplasm cases, it can occur at any age (mean age early adults. fifties) and generally, they present as an adnexal mass, or Some of these cases are inactive in hormonal pattern a haemoperitoneum, with an estrogenic pattern linked to and some are caused by androgen producers [2]. metrorrhagia, endometrial hyperplasia or adenocarci- Macroscopically, sex-cord stromal ovarian tumors are noma. capsulated, vary in size and may be solid or partially cys- These tumors show a differentiation in sex cordons or tic; the cut surface may be grey-white or yellow, de- in specialized stroma: the latter includes female cellular pending on lipid contents; moreover, necrosis and haem- type (granulosa or theca cell), male cellular type (Sertoli orrhage are often present inside the capsule, with cystic or Leydig cells) and undifferentiated elements. compartments filled with fluid or clotted blood. Cellular variety of sex-cord stromal tumor shows The microscopic features of sex-cord stromal ovarian various degrees of differentiation, reproducing a prolifer- tumors may be extremely changeable, even in the same ating pattern present in embryogenesis [1]. tumor, with a wide variety of patterns and characteristic 2. Pathological Findings of Ovarian Call-Exner bodies may be present; they can have in some cases a microfollicular or macrofollicular growth, and in Sex-Cord Stromal Tumors (SCST) other cases trabecular or insular. Granulosa cell tumors contain two histotypes: the young In microscopic diagnosis, cellular elements are often type and the adult type; the latter arises frequently after in the shape of coffee beans and, at an immunohysto-

Copyright © 2010 SciRes JCT 32 Ovarian Sex Cord-Stromal Tumors in Postmenopausal Women and Total Laparoscopical Management chemical level sex-cord stromal ovarian tumors express: estradiol or vimentin, inhibin, CD 99 and cytocheratin, with a reactivity to S100 marker in 50% of cases [3,4]. Currently, the gold standard treatment for SCST is by surgery, even if its surgical management for early-stage adult patients is not well defined. 3. Short Report of an Early SCST in Post-Menopausal Women Managed by Laparoscopy A 69 year-old woman, para 1 (spontaneous delivery) with an uneventful old gynaecological history and menopause at 51 year’s age, was admitted to hospital for geni- tal bleeding and increasing pelvic pain. Clinical exami- Figture 1. Pelvic-abdominal laparoscopical inspection shows nation vaginam exam revealed a palpable right adnexal a large right ovarian cyst, with regular white surface, hold- ing Douglas pouch, occluding visualization of the left ovary mass. Transvaginal ultrasound examination revealed so- and uterus nomorphological uterine fibromatosis with increased endometrial thickness, and a right ovary mass of 8 cm in mally invasive treatment (previously signed on informed diameter, on average, with anechogenic and hypovascu- consent), right ovariectomy was performed by surgeons larized structure, adherent to uterine posterior wall. The using a spill-proof endoscopic device (Endobag, Ethicon ultrasound exam was confirmed by the Computer Tomo- Endo-Surgery, Johnson & Johnson Company, USA). In graphy of pelvis and abdomen, performed in contrast, detail, lysis of right ovarian cysts adhesions was per- which reported no other diseases or anomalies. Bio- formed by aqua dissection with an irrigating probe and, chemical markers for epithelial or germ cell ovarian tu- after this, adnexectomy was carried out with a bipolar mors did not reveal any anomalies and patients wanted a forceps that coagulated the meso-ovarian vessels, the minimally invasive treatment by laparoscopy. Thus, ba- uterus-ovarian and infundibolopelvic ligaments, while sed on the patient’s clinical symptoms and the instru- the ovary was sustained by two Manhes forceps. The mental examinations, diagnostic laparoscopy was per- coagulated ligaments were cut with a monopolar en- formed, following signed detailed informed consent, doscissor and lastly, the tubal wound was irrigated and based on presurgical findings to rule out malignancy. the haemostasis was re-obtained with fine bipolar scis- Laparoscopic access was performed using Direct Visual sors. The excised cyst, emptied in Endobag through a Access method, at the level of umbilicus, with a 10 mm laparoscopic sucking-up pin, was removed via one of the diameter optical trocar (Endopath Xcel Bladeless, Ethi- trocar through a spill-proof endoscopic device, to avoid con Endo-Surgery, Johnson & Johnson Company, USA) accidental leakage of the ovarian cyst content into the inserted through an intra umbilical vertical incision. Fol- peritoneal cavity, which can worsen the prognosis if the lowing the application of carbon dioxide pneumoperito- cyst turns out to be malignant or can lead to peritonitis if neum, with intra-abdominal pressure maintained at 15 the cyst is a dermoid or mucinous cyst. Following laparo- mmHg, the abdominal cavity was inspected through a scopy, Dilatation & Courretage was performed, due to zero-degree laparoscope (Karl Storz, Tuttlingen, Ger- her history of bleeding. The surgical procedure lasted 20 minutes and intra-laparoscopic blood loss was less than many), connected to a video monitor; three supra-pubic 80 cc. ancillary trocars were placed as follows: one 10 mm of The frozen section histology suggested an ovarian diameter trocar inserted in the midline, 3 cm under the SCST, possibly a granulosa cell tumor, while macro- umbilicus, and one in each iliac fossa (5 mm of diameter scopic examinations showed a smooth right ovarian cor- on the left side and 5 mm of diameter on the right size), tical surface. Following pathologists’ advice, in agree- laterally to inferior epigastric vessels. Before both opera- ment with the patient’s preliminary decision and based tive procedures, all pelvic structures were inspected and on her advanced postmenopausal status and slow tumor the abdomen was explored through the laparoscope growth, the surgeons performed a total laparoscopic hys- clockwise. The pelvic-abdominal inspection showed a terectomy (TLH) with left salpingo-ophorectomy (SO), right ovarian cyst, with a regular surface (Figure 1), which avoided radical treatment by surgical staging, lym- some filmy adhesions in the pouch of Douglas and be- phadenectomy, appendectomy and omentectomy. tween the posterior uterine wall and right ovary and tube. Preliminarily, the authors inserted in uterus a trans- The uterus had multiple fibroids and the left ovary was vaginal uterine manipulator (Endopath, Ethicon Endo- normal; following the patient’s decision to receive mini- surgery, Inc., Cincinnati, USA) to quickly mobilize the

Copyright © 2010 SciRes JCT Ovarian Sex Cord-Stromal Tumors in Postmenopausal Women and Total Laparoscopical Management 33 uterus during the procedure and activated PK System LYONS Dissecting Forceps and PK System LP Scissors (Gyrus Medical, Inc. Maple Grove, MN, USA). The left broad ligament anterior leaf was opened while the nurse pushed the uterus into retro-version; the left broad ligament posterior leaf was opened to allow the utero-ovarian ligament and tubes to be defined. This step was performed with PK System LYONS dissecting coagulating forceps and PK System LP monopolar scissors, followed by dissection of the same broad ligament. The uterus was pushed by the nurse towards the opposite side, to the right, using uterine manipulator traction; the left uterine vessels were isolated, the left utero-vesical pouch peritoneum was opened by dissecting and grasping coagulating forceps and the bladder was delicately pushed down- wards. The surgeons completed the uterine devasculari- Figture 2. Histology shows adult-type sex cord-stromal tu- zation procedure, after dissecting the left broad ligament: mors of right ovary; the inset shows tumor cells with mi- they closed the uterine pedicle with a 0 Vicryl, using a crofollicular and coffee bean features (H&E, ×20, original knot pusher after inserting a needle round the uterine magnification) vessels. The left uterine vessels were ligated in two places and incised with scissors between the knots, with vanced malignancy suspect of ovarian cancer; after the final haemostasis of the uterine vessels through bipolar frozen section results, surgeons decided the total laparo- forceps, after checking first the position of the ureters. scopical non-radical unstaged approach either for the The same actions were performed on the right side of patient’s advanced postmenopausal age, or for the biol- uterus, where the ophorectomy had been previously done. ogy of granulosa ovarian mass-a tumor at slow growing. Intra-fascial hysterectomy was continued using the same The morbidity and morbility of a radical laparoscopic dissecting coagulating forceps and monopolar scissors, approach, with pelvic and para-aortic lymphadenectomy, coagulating the cervicovaginal vessels and opening the appendectomy and omentectomy, collide with the patient vagina; anterior and posterior vaginal walls were incised wish and with the stage and the biology of the tumor, with monopolar scissors and haemostasis was completed since more than 95% of SCST are in first stage, with a through coagulating forceps. After uterine extraction, mitotic index less than 3 mitosis for 10 HPF in 75% of surgeons performed laparoscopic suturing of the vaginal cases [5]. cuff. The hysterectomy took 65 minutes and the intra- A recent study by Brown et al. affirm that lymph node laparoscopic blood loss was 200 cc. The final examina- metastasis in ovarian SCST are rare [6]. tion highlighted: a granulosa cell tumor of right ovary The stage of disease is the most important prognostic (Figure 2); the uterus showed fibroids and a cystic glan- factor associated with the risk of relapse: the 5-year sur- dular hyperplasia of endometrium, the left ovary was vival rate of stage I ovarian SCST is 86-96% and all atrophic. other stages are between 26-46% [7]. The FIGO Staging of SCST was IA. The histochemi- There have been only a few case reports on distant cal analysis showed positive reaction to estrogen recep- metastases from these tumors [8,9]. tors, with a proliferating index (KI-67-clone MIB1) of Nevertheless, adult types have a better prognosis than 8% in neoplastic cells. Postoperative recovery was nor- young types, and trabecular and follicular histological mal, the patient was regularly discharged 3 days after pattern has a better prognosis than sarcomatoid pattern. laparoscopy, with no additional therapy at remission and Generally, bilateral involvement of the ovaries in SCST has been followed up for the past three years with in- is observed in only 2% of cases, the opposite ovary can strumental and biochemical monitoring. be preserved in younger women, while, if the uterus 4. Discussion needs to be preserved, endometrium biopsy should be performed because the synchronous occurrence of en- In this patient, the surgeons performed a minimally invasive dometrial adenocarcinoma is associated with estrogen treatment of SCST, avoiding radical treatment by surgical secretion [10]. staging with pelvic and para-aortic lymphadenectomy, The differential histological diagnosis of ovarian appendectomy and omentectomy, for the following rea- SCST is with: undifferentiated carcinoma, small cell car- sons: cinoma, endometrial stroma sarcoma, cellular fibroid, The woman had agreed on minimally invasive treat- endometrioid carcinoma with a sex cord-like growing ment, since presurgical assessment did not show ad- pattern, gonadoblastoma, melanoma or malignant mam-

Copyright © 2010 SciRes JCT 34 Ovarian Sex Cord-Stromal Tumors in Postmenopausal Women and Total Laparoscopical Management malian lobular tumor metastasis [2,3]. 5. Conclusions The SCST prognosis depends on clinical stage, dimen- sion, cellular atipia presence and tumor rupture [11]. Generally, SCST has been treated radically in young As for recurrence, ovarian SCST can recur long after women [15]; since there are no clear conclusions regard- initial treatment and locally, with an average interval of ing the role of postoperative chemotherapy or radiother- 5-10 years, with rare metastasis at distance; the longest apy in stage I patients and in those with complete re- reported interval is 37 years, therefore lifelong follow-up sected tumor [16], TLH plus BSO not followed by lym- care is needed [12]. phadenectomy, has also proven, in our report, to be a safe Factors reported to be associated with outcome include: and effective alternative in postmenopausal patients, with presentation stage, age older than 40, tumor size, tumor fast recovery and minimal morbidity and morbility. In rupture, histologic pattern, high mitotic count, and nu- view of all these findings and due to the technical ad- clear atypicality. vances in endoscopic surgery, laparoscopic non-radical The SCST recurrences are peritoneal and retro-peri- treatment of SCST could be considered a safer treatment toneal masses, histologically well expressed [13]. in selected and consenting adult age patients, which Patients with recurrent disease or residual disease after agrees with a minimally invasive surgery and with a suc- surgery should be treated with a combination of bleomy- cessive prolonged surveillance, also defined as “wait and cin, etoposide, and cisplatin (BEP), while no evidence see” approach, since this tumor grows slowly. shows that treatment with progesterone is beneficial and the role of the radiation therapy, especially for palliation REFERENCES in recurrent disease in the pelvis, has yet to be proven [1] D. Hasiakos, K. Papakonstantinou, E. Karvouni, and S. [10,12]. Fotiou, “Recurrence of granulosa cell tumor 25 years af- A recent article examined the clinical efficacy of ter initial diagnosis,” Report of a case and review of the bevacizumab with or without concurrent chemotherapy literature, European Journal Gynaecolology Oncology, and evaluated the angiogenic characteristics of these tu- Vol. 29, No. 1, pp. 86–88, 2008. mors; authors concluded that anti-VEGF therapy is [2] Rosai and Ackerman’s surgical pathology, 9th edition, highly effective in patients with granulosa cell tumors Mosby Ed., Edinburgh, pp. 1691–1694, 2004. [13]. [3] P. B. Clement and R. H. Young, “Atlas of Gynecologic For the prolonged surveillance, the tumor marker fol- Surgical Pathology, 2nd edition,” Saunders Elsevier Ed., low-up of ovarian SCST is through inhibin, a glycopro- New York, USA, pp. 386–391, 2008. tein produced by granulosa cells, which may be increased [4] S. Leibl, K. Bodo, M. Gogg-Kammerer, and M Ovarian, in postmenopausal women with mucinous carcinomas “granulosa cell tumors frequently express EGFR (Her-1), [14]. Her-3, and Her-4: An immunohistochemical study,” Gy- Concludingly, this short report differs from traditional necol Oncol, Vol. 101, No. 1, pp. 18–23, 2006. treatments since the authors propose, for the first time, a [5] Y. K. Lee, N. H. Park, J. W. Kim, Y. S. Song, S. B. Kang, minimally invasive treatment in a SCST, for the above and H. P. Lee, “Characteristic of recurrence in adult-type reasons. granulosa cell tumor,” International Journal of Gine- Primarily, the presurgical assessment did not show cological Cancer, Vol. 18, No. 4, pp. 642–647, 2008. advanced malignancy suspect of ovarian cancer and the [6] J. Brown, A. K. Sood, M. T. Deavers, L. Milojevic, and D. frozen section results confirmed it. Then, the biology of M. Gershenson, “Patterns of metastasis in sex cord- granulosa ovarian mass, a tumor at slow growing, con- stromal tumors of the ovary: Can routine staging lymp- firms that lymph node metastasis of SCSTs are rare, so as hadenectomy be omitted?” Gynecologic Oncology, Vol. the early stage of this tumor is the most important prog- 113, No. 1, pp. 86–90, 2009. nostic factor associated with the relapse risk, with a sur- [7] D. M. Gershenson, “Management of early ovarian cancer: vival rate of 86–96% in stage I(5-year). Nevertheless, Germ cell and sex cord-stromal tumors,” Gynecologic adult types have a better prognosis than young types. Oncology, Vol. 55, (3 Pt 2), pp. S62–72, 1994. Finally, the mentioned patient was an elder postmeno- [8] O. Ismi, Y. Vayisoglu, T. Karabacak, and M. Unal, pausal women of 69 years, so the morbidity and morbil- “Supraclavicular metastases from a sex cord stromal tu- ity of a more radical laparoscopic approach collided with mor of the ovary,” Tumori, Vol. 95, No. 2, pp. 254–257, the patient wish and with the stage and the biology of the 2009. tumor. [9] A. Chauhan, P. Dahiya, H. Singh, M. Gill, and S. Verma, Given that the literature is lack of laparoscopical “Isolated breast metastasis from granulosa cell tumor of treatments of SCST in adult age patients, based on our the ovary,” Arch Gynecol Obstet, Vol. 280, No. 6, pp. report, a prospective multicentre trial could be proposed 997–999, December 2009. using non-radical laparoscopy in postmenopausal pa- [10] N. Colombo, G. Parma, V. Zanagnolo, and A. Insinga, tients with ovarian SCST. “Management of ovarian stromal cell tumors,” Journal of

Clinical Oncology, Vol. 25, No. 20, pp. 2944–2951, ceptor alpha in ovarian adult granulosa cell tumors and 2007. Sertoli-Leydig cell tumors,” International Journal of Gy- [11] A. Auranen, J. Sundström, J. Ijäs, and S. Grènman, necological Pathology, Vol. 26, No. 4, pp. 375–382, “Prognostic factors of ovarian granulosa cell tumor: A 2007. study of 35 patients and review of the literature,” Interna- [15] G. Hölscher, C. Anthuber, G. Bastert, A. Burges, D. Mayr, tional Journal of Gynecological Cancer, Vol. 17, No. 5, E. Oberlechner, G. Schubert-Fritschle, S. Sinz, H. Som- pp. 1011–1018, 2007. mer, B. Schmalfeldt, and J. Engel, “Malignant Ovarian [12] S. E. Rha, S. N. Oh, S. E. Jung, Y. J. Lee, A. W. Lee, Tumors' of the Munich Cancer Center. Improve- ment of Byun, and J. Y. Recurrent, “Ovarian granulosa cell tu- survival in sex cord stromal tumors—an observational mors: Clinical and imaging features,” Abdom Imaging, study with more than 25 years follow-up,” Acta Obstetri- Vol. 33, No. 1, pp. 119–125, 2008. cia et Gynecologica Scandinavica, Vol. 88, No. 4, pp. 440–448, 2009. [13] X. Tao, A. K. Sood, M. T. Deavers, K. M. Schmeler, A. M. Nick, R. L. Coleman, L. Milojevic, D. M. Gershenson, [16] G. V. Koukourakis, V. E. Kouloulias, M. J. Koukourakis, and J. Brown, “Anti-angiogenesis therapy with bevaci- G. A. Zacharias, C. Papadimitriou, K. Mystakidou, K. zumab for patients with ovarian granulosa cell tumors,” Pistevou-Gompaki, J. Kouvaris, and A. Gouliamos, Gynecology Oncology, Vol. 114, No. 3, pp. 431–436, “Granulosa cell tumor of the ovary: Tumor review,” Inte- September 2009. grative Cancer Therapies, Vol. 7, No. 3, pp. 204–215, 2008. [14] M. A. Farinola, A. M. Gown, and K. Judson, “Estrogen re-

Sum-Based Meta-Analytical Enrichment of Gene Expression Data to Identify Pathway Signatures of Cancers

Kavishwar Wagholikar1, Prasanna Venkatraman2, Sundararajan Vijayraghavan3§, Chandan Kumar-Sinha2*§

1Interdisciplinary School of Scientific Computing, University of Pune, Pune, India; 2Advanced Centre for Treatment, Research and Education in Cancer, Navi Mumbai, India; 3Centre for Advanced Computing, University of Pune campus, Pune, India; *Current Ad- dress: Michigan Center for Translational Pathology, Department of Pathology, University of Michigan Ann Arbor, MI, USA; §Corresponding authors Email: [email protected], [email protected], [email protected], [email protected]

Received November 13th, 2009; revised January 26th, 2010; accepted January 31st, 2010.

ABSTRACT A new method for analysis of microarray gene expression experiments referred to as Sum-based Meta-analytical En- richment (SME) is proposed in this manuscript. SME is a combined enrichment and meta-analytical approach to infer on the association of gene sets with particular phenotypes. SME allows enrichment to be performed across datasets, which to our knowledge was not earlier possible. As a proof of concept study, this technique is applied to datasets from Oncomine, a publicly available cancer microarray database. The genes that are significantly up-/down-regulated (p-value ≤ 10-4) in various cancer types in Oncomine were listed. These genes were assigned to biological processes using GO annotations. The SME algorithm was applied to identify a list of GO processes most deregulated in 4 major cancer types. For validation we examined whether the processes predicted by SME were already documented in literature. SME method identified several known processes for the 4 cancer types and identified several novel processes which are biologically plausible. Nearly all the pathways identified by SME as common to the 4 cancers were found to contribute to processes which are widely regarded as cancer hallmarks. SME provides an intuitive yet objective ‘process-centric’ interpretation of the ‘gene-centric’ output of individual microarray comparison studies. The methods described here should be applicable in the next-generation sequencing based gene expression analysis as well.

Keywords: Gene Expression, Microarray, Enrichment, Meta-Analysis

1. Introduction of genes measured as either over-expressed or under- expressed in the test samples relative to normal. A vari- Microarray based gene expression profiling studies have ety of approaches have been developed in the last decade led to the identification of hundreds of genes differen- to identify gene expression features specific to different tially expressed across diverse sample sets which has cancer types. Rhodes et al. [1] provide a categorization resulted in the characterization of several biomarkers and of these methods. genes of interest. But the bigger promise of providing an insight into underlying biological phenomena continues Statistics based differential analysis lists individual to be challenging. genes which are differentially expressed in a dataset of The problem considered here is that of whole tran- microarray experiments. Clustering [2,3] was the earliest scriptome gene expression analysis by microarrays, attempt to further analyze this list by grouping genes on where genes corresponding to virtually the whole ge- the basis of similarities in their differential expression. nome (25–30,000 genes) are simultaneously interrogated Both these methods are regarded as standard. In con- for their relative expression in a disease type (or subtype) trast, integrated analysis methods examine data in terms and compared against their normal counterpart. For ex- of cancer signatures from other data or other types of ample, breast cancer transcriptome compared against genomic data. These include methods for meta-analysis normal breast transcriptome. This analysis provides a list and functional enrichment.

Meta-analysis has been defined as analysis of multiple On each dataset in the collection, studies are per- datasets. For instance, comparative meta-profiling [4] formed for differential expression across logical group- can be used to identify signatures of genes commonly ings of samples [14] using Student’s t-test. Study specific activated across datasets. gene p-values (or false discovery rates) [15,16] are con- Functional enrichment attempts to interpret the list of sidered to determine the number of studies which infer differential expressed genes derived from one dataset differential expression of a gene at a particular cut-off. using gene annotations like biological process, molecular The results for this step are readily available in On- function and cellular localizations. These methods give a comine. formal framework for biological interpretation which 2) Calculation of Sum-based Meta-analytical En- was subjective in the clustering methods. These include richment Score (S) for query gene-set over-representation approach (ORA) [5], functional class Genes common to the query gene-set and dataset col- scoring (FCS) [6,7] and gene set enrichment analysis [8,9]. lection are identified. For each of the genes in this com- Some of the other reported methods include those mon set, we consider the number of studies in which the based on theory of partially ordered sets [10], random gene is differentially expressed. The sum of the number forests (LeFe) [11], non-parametric pathway based re- of such studies for all genes in the common set gives the gression models, [12] and more recently, impact analysis SME score (S) of the query gene-set. based on systems biology approach [13]. 3) Calculation of significance of the SME score In this paper a new method is proposed, which in- To evaluate the significance of the score, SME score volves a combined enrichment and meta-analytical com- for a million random gene sets, each having the same parison, to infer on the association of gene sets (for ex- number of genes as the common set (derived in previous ample those constituting Gene Ontology, GO) with the step) is computed. The fraction of the number of random specific phenotypes investigated in a set of microarray gene-sets for which the SME score is greater than or experiments. We refer to the method as Sum-based equal to the statistic for the gene-set in question, gives Meta-analytical Enrichment (SME). SME allows en- the s-value for the association. The term s-value is used richment to be performed across datasets, which to our to differentiate it from the analogous p-value, since the knowledge was not possible earlier. Our premise is that distribution computed in step 2 is a surrogate for the null in order to implicate a ‘process’ (eg. GO Processes) in distribution. The null-distribution cannot be accurately the mechanism of a pathology (eg. cancer), the various determined due to dearth of information about proc- independent ‘events’ (expression changes in individual ess-phenotype associations. genes) constituting and/or contributing to the process serve as surrogates to studying the process itself. Greater 2.1 Mathematical Description of SME the evidence of association of such constituent events A) Input for the method: with a pathology, greater is the confidence in the 'proc- Gi : set of genes in the collection of microarray data- ess-association’ hypothesis. SME is a heuristic extension sets. of this notion. In order to calculate the significance of a Gj : the query gene set gene-set phenotype association, it considers both – 1) the B) Steps: number of genes in the set which are differentially ex- 1) Determination of number of studies reporting pressed in a microarray study, and 2) the number of differential expression of a gene Let Y(g) give the total studies which report each of the genes as differentially number of studies across logical groupings in individual expressed. Permutation testing has been used to rule-out datasets which infer that a gene g is differentially ex- chance associations. pressed on the basis of a p-value/q-value cut-off. As a proof of concept, this technique was applied to a 2) Calculation of SME score (S) for query gene-set publicly available cancer-microarray database (On- comine) to identify Gene Ontology processes dys- SG()j   Yg () regulated in four different cancer types and an assess- gG()ji G ment was carried to examine whether the identified asso- 3) Determination of significance of the score ciations were already reported in literature. With the above formula, S (Gr) is calculated for very 2. Methods large number of random gene sets (Gr) from Gi, such that Given a collection of microarray gene-expression data- |Gr| = |Gj ∩ Gi | sets for a phenotype, the significance of association of a The s-value is given by, query gene-set with the phenotype, is evaluated with fol- PG() X ()/ G n lowing steps: jr rn(1.. ) 1) Determination of number of studies reporting differential expression of a gene where for large number n of Gr ,

 For validation, an assessment was made whether the 1 whenSGrj SG  processes identified for the four major cancer types have XG  r  already been reported to be associated with the particular  0 when S G S G  rj cancer types. Further, the set of processes common to the four cancer types were analyzed. 2.2 Case Study 2.3 Implementation Oncomine is a publicly available cancer microarray da- The algorithm was implemented on a Linux (Fedora core tabase [17]. The SME method was applied to datasets in 6) system using Perl and MySQL database, with the Oncomine (Version 2.0) to identify Gene Ontology permutation testing module implemented in C++. A processes associated with and common to four major web-based front-end for calculating significance of asso- cancer types — breast cancer, prostate cancer, leukemia ciation of a gene-set with a particular cancer type was and lymphoma. Oncomine was queried for all the human based on the discussed algorithm, using apache server genes (listed at NCBI) to obtain the number of studies and Support-Vector-Graphics. The tool was tested for its reporting a gene as differentially expressed for a p-value precision of s-value determination for a set of GO proc- cut-off of 10-4. All of the Gene Ontology biological esses. As seen in Figure 2, overall the standard deviation process terms were listed from the GO database of the s-value was less than 0.0008 and decreased to- (http://www.geneontology.org) [18]. In GO database, wards the extremes, which ensures the accuracy of results. processes are segregated as biological processes, cellular components and/or molecular functions and genes con- 3. Results and Discussion tributing to each process are annotated. For each of the To facilitate the assessment similar GO processes were GO biological processes, the contributing human genes grouped by clustering on basis of the overlap of con- were obtained by querying with the GO association stituent gene sets. “contributes to”. Of the 9561 GO biological processes, In case of leukemia, platelet activation, hemoglobin those which had insufficient information in oncomine biosynthesis and thymic T cell activation could be read- were excluded. This was done by excluding processes ily associated with the physiology and metabolism of which had less than 10 of their contributing genes listed blood cells and lymphocytes. A leukemia-associated in oncomine. Further, processes which have more than CD9 glycoprotein antigen is known to have groups of 50 genes represented in oncomine were excluded to re- N-acetyl glucosamine residues, which may explain the move highly generic processes. 323 processes remained, identification of acetyl glucosamine metabolism [19]. Involvement of sphingolipid metabolism is known in and the association of each process with 4 cancer types hematological malignancies and sphingolipids have been was tested. SME was applied to obtain the s-value of investigated as mediators of apoptosis triggered in re- association of the processes with each of the 4 cancer sponse to anti-leukemic agents [20]. types and the GO processes with s-value ≤ 0.01 (em- As seen for leukemia, several pathway changes listed pirically approximated) were listed. for lymhpoma are also physiologically relevant. Some of For example, to test the association of the GO process these are associated with chemokine (interleukine) and “insulin receptor signaling pathway” (GO: 0008286) chemokine receptor genes. These chemokines and their with breast cancer, a query was performed on GO data- receptors are involved in the development and differen- base to get all the genes annotated to this processes with tiation of immune cells. They have also been found to be the relation “contributes to”. There are 21 such genes from which information is available for 12 genes, in On- comine. Oncomine lists the number of breast cancer studies reporting any of these genes as differentially expressed (with default p-value cut-off of 10-4), in various class comparisons. These are summed (score=12) and this score is used to calculate the s-value of association between “insulin receptor signaling pathway” and “breast cancer” by forming one million random sets of 12 genes and calculating the SME score for the random sets; the number of random gene sets with an SME score equal to or more than 12 is divided by the total number of random gene sets (one million) to get the s-value of 0.000534 (Figure 1). Figure 1. Significance testing

Figure 2. Plot of mean vs. standard deviation was obtained by multiple runs of the tool for a subset of GO processes from the study present in cells from other carcinomas like those of colon. process which is regarded as an hallmark of cancer [31]. The receptors are GTP binding and specific association Regulation (dysregulation) of mitosis which grants of these receptors with lymphoma, a result emerging limitless replicative potential to cancers has been aptly from our study is noteworthy [21–24]. Matrix metallo- identified in our analysis as well. Cell cycle check point proteinases (MMPs) have been implicated in lymphomas genes maintain sensitivity to growth signals in normal [25] and MMPs are known for collagenolysis [26], which cells. JAK-STAT cascade is downstream of many cyto- is identified by SME. kine and growth hormone receptors. The activation of In contrast to both the liquid tumors, the two major transcription factors of the STAT (signal transducer and solid tumor types — breast and prostate showed diverse activator of transcription) protein family by JAK (Janus pathway associations. In the case of breast cancer, small activated kinase) is reported to be constitutively activated GTPase Rho signaling pathways are known to regulate in a many types of cancers [32]. Intracellular receptor breast cancer cells [27]. The human epidermal growth mediated signaling pathways are known to impart car- factor receptor (HER-2) oncogene is known to encode a cinogenic attribute of self sufficiency of growth signals. transmembrane tyrosine kinase receptor which increases Regulation of the actin cytoskeleton is critically involved invasiveness of breast cancer [28]. HER-2 has been tar- in endothelial cell migration required for angiogenesis geted for therapy. In the case of prostate cancer, androgen receptor is [33], which is an important hallmark of cancer. Cellular known to interact with an array of growth factor signal protein catabolism represents the cachexia which is seen transduction events including epidermal growth factor in endstage of cancers and mediators of protein catabo- and vascular endothelial growth factor [29]. Pathways lism have been targeted as cure for cancer cachexia [34]. that involve smooth muscle contraction and sugar trans- Identification of steroid hormone receptor signaling port have been shortlisted. These processes are vital to pathway conforms to the knowledge that steroid group, the normal functions associated with the prostate gland. among all hormones is known to act as growth factor The dominant association of matrix adhesion pathways promoting an array of different cancers [35]. In summary may be due to their role in metastasis [30]. Bone and nearly all the identified common pathways, were found lymph node metastasis are common in prostate cancer. to contribute to processes which are widely regarded as The pathways common to the four cancers (see Figure cancer hallmarks. 3) included NF-kappaB (nuclear factor-kappa-B) Apart from sharing some of the outcomes projected by -mediated survival pathway. This is widely regarded as other studies which have attempted to identify pathway the mechanism by which cancer cells evade apoptosis, a changes, our analysis has several unique advantages due

Figure 3. Pathways common to breast cancer, prostate cancer, leukemia and lymphoma to the analytical method employed and the use of ex- This obviates the analytical problems that are often asso- perimental sets — a) A feature of the method is the large ciated with animal models. scale combined analysis of data which intrinsically pro- 3.1 Limitations vides noise-reduction and highlights the consistent features associated with the phenotype. b) Most importantly One limitation of SME is that extrapolation of the s-value all the data used for analysis are from human samples. of the intersection (of genes annotated to a GO process

Copyright © 2010 SciRes JCT Sum-Based Meta-Analytical Enrichment of Gene Expression Data to Identify Pathway Signatures of Cancers 41 and present in the microarray experiments) to a GO ing,” Nature, Vol. 403, No. 6769, pp. 503–511, 2000. process becomes less reliable when the gap between the [3] L. J. van 't Veer, et al., “Gene expression profiling pre- intersection and the annotation widens, such as for proc- dicts clinical outcome of breast cancer,” Nature, Vol. 415, esses higher in the GO hierarchy. This problem could be No. 6871, pp. 530–536, 2002. circumvented by excluding the processes which have a [4] D. R. Rhodes, et al., “Large-scale meta-analysis of cancer gap greater than an optimal cut-off. Also, the analysis microarray data identifies common transcriptional pro- could be made more stringent with a correction for mul- files of neoplastic transformation and progression,” Pro- tiple hypothesis testing. In the present analysis, we have ceedings of the National Academy of Science, U S A, Vol. avoided both the cut-off for intersection gaps and the 101, No. 25, pp. 9309–9314, 2004. multiple hypothesis correction; because our analysis is [5] P. Khatri and S. Draghici, “Ontological analysis of gene exploratory in nature and processes higher in the GO expression data: Current tools, limitations, and open hierarchy at times help to establish a context for the de- problems,” Bioinformatics, Vol. 21, No. 18, pp. 3587– lineated children. 3595, 2005. [6] J. J. Goeman, et al., “A global test for groups of genes: 4. Conclusions Testing association with a clinical outcome,” Bioinfor- matics, Vol. 20, No. 1, pp. 93–99, 2004. SME method identified several known processes for se- lecting cancer types. Moreover novel processes were [7] P. Pavlidis, et al., “Using the gene ontology for microarray data mining: a comparison of methods and application delineated which are biologically plausible and have po- to age effects in human prefrontal cortex,” Neurochemical tential utility. Nearly all the pathways identified by SME Research, Vol. 29, No. 6, pp. 1213–1222, 2004. as common to different cancers were found to contribute to processes which are widely regarded as cancer hall- [8] V. K. Mootha, et al., “PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately marks. With the accrual of micro-array results in reposi- down regulated in human diabetes,” Nature Genetics, Vol. tories, expansion of GO database and further optimiza- 34, No. 3, pp. 267–273, 2003. tions, the method can be expected to lead to increasingly [9] A. Subramanian, et al., “From the Cover: Gene set en- accurate output. SME makes it possible to draw infer- richment analysis: A knowledge-based approach for in- ences based on a large scale combined analysis of mi- terpreting genome-wide expression profiles,” Proceedings croarray data by reducing noise and has an advantage in of the National Academy of Science, U S A, Vol. 102, No. its intuitive yet objective approach. 43, pp. 15545–15550, 2005. 5. Authors’ Contributions [10] R. Barriot, D. J. Sherman, and I. Dutour, “How to decide which are the most pertinent overly-represented features KBW and CKS developed the algorithm and conceived during gene set enrichment analysis,” BMC Bioinformat- the design of the case study. ics, Vol. 8, pp. 332, 2007. KBW and VS formulated the algorithm and imple- [11] G. S. Eichler, et al., “The LeFE algorithm: Embracing the mented the case study. They also conceived and con- complexity of gene expression in the interpretation of mi- ducted the automated evaluation. croarray data,” Genome Biology, Vol. 8, No. 9, pp. R187, PV and CKS assessed the utility of the method to pro- 2007. vide biological insight into the case study. [12] Z. Wei and H. Li, “Nonparametric pathway-based regres- All authors wrote the manuscript and approved it. sion models for analysis of genomic data,” Biostatistics, Vol. 8, No. 2, pp. 265-284, 2007. 6. Acknowledgements [13] S. Draghici, et al., “A systems biology approach for pathway level analysis,” Genome Research, Vol. 17, No. 10, We are thankful for the developers of Gene Ontology, pp. 1537–1545, 2007. Oncomine, GoPubMed and NCBI for making their data/tools available for free public access. The technical [14] D. R. Rhodes, et al., “Oncomine 3.0: genes, pathways, and networks in a collection of 18,000 cancer gene ex- assistance given by Chitra Alavani (ISSC, University of pression profiles,” Neoplasia, Vol. 9, No. 2, pp. 166–180, Pune) is duly acknowledged. We thank Dr. Gadre (Di- 2007. rector, ISSC) for making computational resources avail- [15] J. Quackenbush, “Computational analysis of microarray able for this study. data,” Nature Reviews Genetics, Vol. 2, No. 6, pp. 418– 427, 2001. REFERENCES [16] D. Stekel, Microarray Bioinforamtics, Cambridge Uni- [1] D. R. Rhodes and A. M. Chinnaiyan, “Integrative analysis versity Press, Cambridge, 2003. of the cancer transcriptome,” Nature Genetics, supple- [17] Oncomine, Available from: http://www.oncomine.org. 2007. ment Vol. 37, pp. S31–S37, 2005. [18] M. A. Harris, et al., “The Gene Ontology database and [2] A. A. Alizadeh, et al., “Distinct types of diffuse large informatics resource,” Nucleic Acids Research, Vol. 32 B-cell lymphoma identified by gene expression profil- (Database issue), pp. D258–D261, 2004.

[19] Y. Komada and M. Sakurai, “Shedding of CD9 antigen in phoma, Vol. 39, No. 5–6, pp. 485–493, 2000. acute lymphoblastic leukemia,” Leukemia and Lym- [27] Y. Tang, et al., “Role of Rho GTPases in breast cancer,” phoma, Vol. 12, No. 5-6, pp. 365–372, 1994. Frontiers in Bioscience, Vol. 13, pp. 759–776, 2008. [20] C. Ricci, F. Onida, and R. Ghidoni, “Sphingolipid players [28] J. S. Ross, et al., “The HER-2 receptor and breast cancer: in the leukemia arena,” Biochimica et Biophysica Acta, Ten years of targeted anti-HER-2 therapy and personal- Vol. 1758, No. 12, pp. 2121–2132, 2006. ized medicine,” Oncologist, Vol. 14, No. 4, pp. 320–368, [21] R. Huang, A. Wallqvist, and D. G. Covell, “Targeting 2009. changes in cancer: Assessing pathway stability by com- [29] M. L. Zhu and N. Kyprianou, “Androgen receptor and paring pathway gene expression coherence levels in tu- growth factor signaling cross-talk in prostate cancer mor and normal tissues,” Molecular Cancer Therapeutics, cells,” Endocrine-Related Cancer, Vol. 15, No. 4, pp. 841 Vol. 5, No. 9, pp. 2417–2427, 2006. –849, 2008. [22] H. Kulbe, et al., “The chemokine network in can- [30] S. Varambally, et al., “Integrative genomic and proteomic cer--much more than directing cell movement,” Interna- analysis of prostate cancer reveals signatures of metas- tional Journal of Developmental Biology, Vol. 48, No. tatic progression,” Cancer Cell, Vol. 8, No. 5, pp. 393– 5-6, pp. 489–496, 2004. 406, 2005. [23] J. Meijer, et al., “The CXCR5 chemokine receptor is [31] D. Hanahan and R. A. Weinberg, “The hallmarks of can- expressed by carcinoma cells and promotes growth of cer,” Cell, Vol. 100, No. 1, pp. 57–70, 2000. colon carcinoma in the liver,” Cancer Research, Vol. 66, No. 19, pp. 9576–9582, 2006. [32] A. C. Pfeifer, J. Timmer and U. Klingmuller, “Systems biology of JAK/STAT signaling,” Essays in Biochemistry, [24] G. Opdenakker and J. Van Damme, “The countercurrent Vol. 45, pp. 109–120, 2008. principle in invasion and metastasis of cancer cells. Re- cent insights on the roles of chemokines,” International [33] J. T. Durham and I. M. Herman, “Systems biology of Journal of Developmental Biology, Vol. 48, No. 5-6, pp. JAK/STAT signalling: Inhibition of angiogenesis in vitro: 519–527, 2004. a central role for beta-actin dependent cytoskeletal re- modeling,” Microvascular Research, Vol. 45, No. 3, pp. [25] J. L. Lauer-Fields, D. Juska, and G. B. Fields, “Matrix 109–120, 2008. metalloproteinases and collagen catabolism,” Biopoly- mers, Vol. 66, No. 1, pp. 19–32, 2002. [34] J. M. Argilés, et al., “Catabolic mediators as targets for cancer cachexia,” Drug Discovery Today, Vol. 8, No. 18, [26] A. E. Kossakowska, S. J. Urbanski, and A. Janowska- pp. 838–844. 2003. Wieczorek, “Matrix metalloproteinases and their tissue inhibitors-expression, role and regulation in human ma- [35] S. Cheng and S. Balk, Steroid Hormone Receptor Signal- lignant non-Hodgkin's lymphomas,” Leukemia and Lym- ing in Cancer.

Role of Estradiol, Progestins, Insulines and Adipocytokines in Breast Cancer Promotion in Post-Menopausal Women

Christian Jamin

AFACS 169 Bd Haussmann, Paris, France. Email: [email protected]

Received December 28th, 2009; revised January 22nd, 2010; accepted February 3rd, 2010.

ABSTRACT Estrogens and artificial progestins used in hormone replacement therapy increase breast cancer risk. This seems to be due to a promoting and not initiating effect. A synergic effect of estradiol and hyperinsulinism has been shown. Insulin plays a role in the increase of breast cancer risk when associated with android obesity, sedentariness, type II diabetes, and high glycemic index food, alcohol and trans fatty acids intake. Natural menopause induces insulin resistance and does not induce a risk decrease. The role of insulin gives a new outlook on the influence of HRT in breast cancer promotion: estradiol alone, which improves insulin-sensitivity, does not increase breast cancer risk. Artificial progestins associated with estrogens increase the risk, whereas estrogens associated with progesterone do not. This could be due to the fact that artificial progestins increase insulin resistance, whereas natural progesterone does not. Adipose tissue, which is an endocrine gland, is insulin dependant. Breast cancer and its seriousness are correlated to adipocytokin circulating levels such as resistin, leptin, interleukin 1, adipocyte fatty acid-binding protein, and are inversely correlated to the level of adiponectin. Insulin could play a synergic role with sexual steroids by a direct effect and by increasing adipose tissue secretions.

Keywords: Breast Cancer, Estrogens, Progestins, Insulin, Hormone Replacement Therapy, Adipocytokin

1. Introduction 1,26 RR [3]. The Women Health Initiative (WHI) study confirms this increase after 5 years of treatment [4], but The fact that breast cancer mainly affects women after only among women taking HRT before the study. It is puberty and regresses after an ovariectomy can logically thus impossible to evaluate the necessary HRT duration lead to the conclusion that sexual steroids play a pre- before a significant risk increase appears. The Million dominant part in the genesis and the promotion of breast Women Study (MWS) [5] is in accordance with the Lan- cancer. For numerous researchers, this predominant part cet meta-analysis and WHI. Only women with a BMI has become a unique part, and consequently they have higher than 25 kg/m² have a risk increase [4]. come to make estrogens a scapegoat in all epidemiologi- Globally RR is low, but since spontaneous risk in- cal and clinical observations. A few disturbing elements creases with age, this low RR could induce a high attrib- like the fact that breast cancer incidence still rises after utable risk over long treatment periods (10 or 15 years) menopause, whereas it decreases sharply after hormone [6]. When one considers the rapidity at which excess risk replacement therapy (HRT) has been stopped, raise some appears, plus the fact that this risk increase disappears questions about the validity of such an approach [1,2]. A new rapidly when the treatment is stopped, plus the absence of actor, insulin, might play a predominant part in the future. in situ cancers increase, it is likely that there is a promo- 2. Role of Sexual Steroids in the Initiation tion phenomenon and not of initiation [1,2]. The arm and/or Promotion of Post Menopausal estrogens only of WHI, as well as numerous other studies Breast Cancer: State of the Art of the cohort, do not show any increase with estrogens only (RR=0,77 (0,57–1,06) [7,8]. One study with a very A Lancet meta-analysis compiled studies on the relation- long follow-up is the only one to show a very late risk ship between HRT and breast cancer risk and found a increase (more than 15 years) with estrogens only [9].

The French E3N cohort study does not show any risk prognosis. increase with the association estrogens-natural proges- Insulin stimulates normal and cancer cell proliferation terone or retroprogesterone and confirms the risk increase and has a promoting effect on breast tumours in the ani- when estrogens and synthesis progestins are associated, mal. The insulin receptor is over expressed in breast can- whatever their type [10]. All these results are concordant cer, and in woman hyperinsulinism is associated with an with the measure of apoptosis proliferation which shows increase in estrogen ovarian production, a decrease of a maximum promotion with the association estrogens- estradiol carrier protein (SHBG), and therefore an in- MPA or NETA, an intermediate promotion with estro- crease of free estradiol [15,24–29]. The fixation of insulin gens only and a minimum promotion with the association to its receptor on breast cancer cell increases phosphati- estrogens-progesterone and retroprogesterone or tibolone dylinositolkynase and kinase MAP activity. Furthermore, [11]. A randomized study versus placebo shows a sig- insulin activates estrogens receptor alpha-mediated tran- nificant breast cancer risk decrease after 3 years of ti- scription in breast cancer cell lines, even in the absence bolone [12]. Cancers discovered during HRT have a bet- of estradiol. In addition estradiol activates the MAPK ter prognosis, which could be due to a better differenti- pathway of insulin. Finally, in breast cancer cell culture ated histologic form. They are almost exclusively lobular medium, estradiol has a proliferative effect only in the or lobulo ductal cancers, with more hormone-sensitive presence of growth factors such as insulin. forms (E3N) [13–15]. Their metastatic risk evaluated This throws a different light over numerous situations over 20 years is weaker, whatever the site [16]. These that could not be explained until now. With menopause favourable prognosis characteristics are observed only comes insulin resistance: this could explain the fact that among women treated with estrogens associated with breast cancers do not diminish at menopause, but do di- progestins and not among women treated with estrogens minish rapidly when HRT is stopped [1,3,30]. Over- weight is a recognised breast cancer risk factor after only or with the association estrogens/progesterone. menopause, whereas it is rather protective before meno- 3. The Role of Hyperinsulinism in Breast pause. In this regard we know that pre-menopausal gy- Cancer noïd obesity does not alter insulin-sensitivity, whereas post-menopausal android obesity is associated with insu- In the observational WHI study, each woman entering the lin-resistance [31]. study had an exhaustive hormonal biological testing. The Metabolic syndrome and its clinical marker, waist- correlations between the hormone levels of women when to-hip ratio, multiplies by 2 breast cancer risk, and this they entered the study and their subsequent breast cancer independently from testosterone level. Furthermore this risk have shown that there is not one but two actors very ratio is predictive of breast cancer mortality at the time predictive of this risk: blood levels of estradiol and insu- cancer is discovered [32–35]. lin [17]. It has to be pointed out that in this study neither The level of HDL cholesterol, a marker of insulin sen- IGF1 nor IGFBP3 are independent risk factors. The insu- sitivity, is inversely correlated to insulinemia and breast lin effect is not affected by the adjustment of estradiol cancer risk [35,36]. level, and similarly the adjustment of insulin level does Nowadays physical activity is admitted to be a protec- not affect the effect of estradiol. Therefore these two fac- tive factor from breast cancer, as well as a reduction factors are independent. As of today, three others studies tor of breast cancer mortality [18,37,38]. This protection have evaluated in a prospective way the insulin level as a also exists among women with a BRCA1/2 mutation [39]. predictive breast cancer risk factor. Two of them do not In increasing muscular mass, physical activity improves find any correlation, but they included women under insulin-sensitivity and thus reduces circulating insulin HRT. The last one, which did not include women under level, including after breast cancer [40]. HRT, finds the same result as observational WHI. An- High trans fatty intakes are associated with a high other study showed that when a post-menopausal woman breast cancer risk RR=1,75 (1,08–2,83) in E3N , and with has breast cancer, her insulin level at the time of diagno- a higher insulin resistance risk and diabetes [41,42]. Al- sis is strongly predictive of her mortality risk after 10 cohol, which is a recognized breast cancer risk factor years [18–21]. after menopause, reduces insulin-sensitivity and increases In another study, after stratification of insulinemia in insulin receptors in tumours [43]. Food glycemic index is quartiles, it has been found that the insulinemia superior a marker of insulinic response to their ingestion. The quartile versus the inferior quartile gives a mortality RR more food with high glycemic index a woman eats, the at 8 and of recurrence at 4 [22]. In the MA 14 study, in- higher her breast cancer risk [44,45]. sulin resistance is associated with survival reduction Type II diabetics, non insulin-dependent, have a hy- without recurrence [23]. Finally a preliminary work perinsulinism and a 1,2 breast cancer relative risk. Stud- shows that when insulin receptors are measured in tu- ies have shown that when these type II diabetes are mours, tumours with the most receptors have a worse treated with metformin, there is a reduction of breast

Copyright © 2010 SciRes JCT Role of Estradiol, Progestins, Insulines and Adipocytokines in Breast Cancer Promotion in Post-menopausal Women 45 cancer mortality, whereas a treatment by sulfonyl ureas node invasion. It increases mammary cancer cell apop- or by insulin increases this mortality [46–48]. tosis and diminishes neo-vascularization [51–54]. 4. Interactions between Insulin Sensitivity 6. Conclusions and HRT Thus the insulin/estrogens couple has a synergic and in- This estradiol/insulin synergy enlightens the relationship dependent effect on breast cancer risk. Estradiol does not between HRT and breast cancer risk. We know that es- promote these cancers in the absence of hyperinsulinism. trogens given at mean doses by oral or transcutaneous Insulin-dependent adipose tissue could also be one of the routes improve insulin-sensitivity and thus diminish insu- actors of this synergy, through adipocytokines. linemia, whereas oral estrogens given at high doses This approach gives new elements for a better under- and/or associated with artificial progestin, in particular standing of the relationship between hormones and breast MPA, increase insulin-resistance [30]. Furthermore the cancer, and might in the future originates in new thera- PEPI study has shown that contrary to MPA, natural peutic and/or preventive strategies in breast cancer. progesterone does not alter insulin-sensitivity, which is improved by equine conjugated estrogens [49]. Thus it REFERENCES could be because of insulin that HRT containing artificial [1] R. T. Chlebowski., L. H. Kuller, R. L. Prentice, M. L. progestins increase breast cancer risk, that estrogens only Stefanick, and E. Manson, “Breast cancer after use of es- do not increase or maybe even reduce this risk, and that, trogen plus progestin in post menopausal women,” New contrary to progestins, progesterone associated with es- England Journal of Medicine, Vol. 360, pp. 573–587, trogens have no deleterious effect [7,10]. 2009. Moreover, deleterious effects of HRT containing arti- [2] P. M. Ravdin, K. A. Cronin, N. Howlader, C. D. Berg, ficial progestin are maximum in immediate post- and R. T. Chlebowski, “The decrease in breast-cancer in- menopause when insulin resistance has not set in. At dis- cidence in 2003 in the United States,” New England tance from menopause, the same treatments have a Journal of Medicine, Vol. 356, pp. 1670–1674, 2007. weaker effect because insulin resistance linked to meno- [3] Collaborative Group on Hormonal Factor in Breast Can- pause has already had its own effect [50]. cer, “Breast cancer and hormone replacement therapy: Collaborative reanalysis of data from 51 epidemiological 5. Indirect Effects of Insulin on Breast studies of 52705 women with breast cancer and 108411 Cancer Risk: Role of Adipose Tissue women without breast cancer,” Lancet, Vol. 350, pp. 1042–1059, 1997. and Adipocytokines [4] J. E. Rossouw, G. L. Anderson, R. L. Prentice, A. Z. La- Adipose tissue secretes hormones called adipocytokines. Croix, and C. Kooperberg, Writing group for the WHI The volume of adipose tissue is increased by insulin. “Risks and benefits of estrogen plus progestin in healthy Thus insulin could play a direct part on cancerous cells, postmenopausal women: principal results,” From the Women's Health Initiative randomized controlled trial, but could also have an indirect effect in modifying adi- Journal of the American Medical Association, Vol. 288, pocytokines secretions. Among these adipocytokines, pp. 321–333, 2002. resistin is higher in women with breast cancer, is associ- [5] V. Beral, Million Women Study Collaborators, “Breast ated with cancer seriousness and is partly responsible for cancer and hormone-replacement therapy in the Million insulin-resistance. Leptin, another adipocytokine, is also Women Study,” Lancet, Vol. 362, pp. 419–427, 2003. high among women with breast cancer, metastatic in par- [6] M. Ewertz, L. Mellemkjaer, H. Poulsen, S. Friis, and H. T. ticular, and increases aromatase activity in intra- Sørensen, “Hormone use for menopausal symptoms and mammary production of estrogens. Interleukin 1, also risk of breast cancer. A Danish cohort study,” British secreted by adipose tissue, has a level correlated to breast Journal of Cancer, Vol. 92, pp. 1293–1297, 2005. cancer risk and progression. Adipocyte fatty acid binding [7] G. L. Anderson, M. Limacher, A. R. Assaf, T. Bassford, protein is predominantly expressed in the cytosol of ma- and S. A. Beresford, “Effects of conjugated equine estro- ture adipocytes and has been described as associated with gen in postmenopausal women with hysterectomy. The obesity markers and obesity-related diseases. Adipocyte Women's Health Initiative randomized controlled trial,” fatty acid binding protein is recognized to affect insulin Journal of the American Medical Association, Vol. 291, sensitivity, lipid metabolism, and an inflammatory re- pp.1701–1712, 2004. sponse associated with atherosclerosis. Adipocyte fatty [8] W. Y. Chen, J. E. Manson, S. E. Hankinson, B. Rosner, acid binding protein has been recently shown to be sig- and M. D. Holmes, “Unopposed estrogen therapy and the nificantly associated with breast cancer risk. Finally adi- risk of invasive breast cancer,” Archives of Internal ponectin is lowered in case of metabolic syndrome, type Medicine, Vol. 166, pp. 1027–1032, 2006. II diabetes and android obesity. It is low in women with [9] C. I. Li, K. E. Malone, P. L. Porter, N. S. Weiss, and M. T. breast cancer and its level is inversely associated with Tang, “Relationship between long durations and different

regimens of hormone therapy and risk of breast cancer,” A. R. Folsom, “Serum insulin and glucose levels and Journal of the American Medical Association, Vol. 289, breast cancer incidence,” American Journal of Epidemi- pp. 3254–3263, 2003. ology, Vol. 156, pp. 349–352, 2002. [10] A. Fournier, F. Berrino, E. Riboli, V. Avenel, and F. [22] P. Goodwin, M. E. Ennis, K. Pritchard, M. Trudeau, and J. Clavel-Chapelon, “Breast cancer risk in relation to types Koo, “Fasting insulin and out come in early-stage breast of hormone replacement therapy in the E3N-EPIC co- cancer: Results of a prospective cohort study,” Journal of hort,” International Journal of Cancer, Vol. 114, pp. Clinical Oncology, Vol. 20, pp. 42–51, 2001. 448–454, 2005. [23] M. N. Pollak, J. W. Chapman, L. Shepered, D. Meng, and [11] C. E. Wood, T. C. Register, C. J. Lees, H. Chen, and S. P. Richardson, “Insuline resistance, estimed by serum Kimrey, “Effects of estradiol with micronized progester- C-peptide level, is associated with reduced event-free sur- one or medroxyprogesterone acetate on risk markers for vival for post menopausal women in NCIC CGT MA.14 breast cancer in postmenopausal monkeys,” International adjuvant-breast cancer trial,” Journal of Clinical Oncol- Journal of Cancer, Vol. 101, pp. 125–134, 2007. ogy, Vol. 24, pp. 524, 2006. [12] S. Cummings, B. Ettinger, P. Delmas, P. Kenemans, and [24] D. Ish-Shalom, C. T. Christoffersen, P. Vorwerk, P., N. V. Stathopoulos, “The effects of Tibolone in older post- Sacerdoti-Sierra, and R. M. Shymko, “Mitogenic proper- menopausal women,” New England Journal of Medicine, ties of insulin and insulin analogues mediated by the insu- Vol. 359, pp. 697–708, 2008. lin receptor,” Diabetologia, Vol. 40, Supplement No. 2, [13] W. Chen, D. Pettiti, and A. Geiger, “Mortality following pp. S25–31, 1997. development of breast cancer while using oestrogen or [25] L. Poretsky and M. F. Kalin, “The gonadotropic function oestrogen plus progestin: A computer record-linkage of insulin,” Endocrine Reviews, Vol. 8, No. 132–141, study,” British Journal of Cancer, Vol. 93, pp. 392–398, 1987. 2005. [26] S. M. Shafie and F. H. Grantham,“Role of hormones in [14] A. Fournier, F. Berrino, and F. Clavel-Chapelon, “Un- the growth and regression of human breast cancer cells equal risks for breast cancer associated with different (MCF-7) transplanted into athymic nude mice,” Journal of hormone replacement therapies: Results from the E3N the National Cancer Institute, Vol. 67, pp. 51–56, 1981. cohort study,” British Journal of Cancer, Vol. 107, pp. [27] M. Pugeat, J. C. Crave, M. Elmidani, M. H. Nicolas, and 103–111, 2008. M. Garoscio-Cholet, “Pathophysiology of sex hormone [15] C. Schairer, M. Gail, C. Byrne, P. S. Rosenberg, and S. R. binding globulin (SHBG): Relation to insulin,” Journal of Sturgeon, “Estrogen replacement therapy and breast can- Steroid Biochemistry and Molecular Biology, Vol. 40, pp. cer survival in a large screening study,” Journal of the Na- 841–849, 1991. tional Cancer Institute, Vol. 91, pp. 264–270, 1999. [28] A. Belfiore, “The role of insulin receptor isoforms and [16] F. Schuetz, I. J. Diel, M. Pueschel, T. von Holst, and E. F. hybrid insulin/IGF-1 receptors in human cancer,” Current Solomayer, “Reduced incidence of distant metastases and Pharmaceutical Design, Vol. 13, pp. 671–686, 2007. lower mortality in 1072 patients with breast cancer with a [29] J. Chappell, W. Leitner, S. Solomon, I. Golovchenko, and history of hormone replacement therapy,” American M. Goalstone, “Effect of insulin on cell cycle progression Journal of Obstetrics and Gynecology, Vol. 196, pp. 342. in MCF-7 breast cancer cells,” Journal of Biological e1-9, 2007. Chemistry, Vol. 276, pp. 38023–38028, 2001. [17] M. J. Gunter, D. R. Hoover, H. Yu, S. Wassertheil- [30] S. R. Lindheim, S. C. Presser, E. C. Ditkoff, M. A. Vijod, Smoller, and T. E. Rohan, “Insulin, insulin-like growth and F. Z. Stanczyk, “A possible bimodal effect of estro- factor-1, and risk of breast cancer in postmenopausal gen on insulin sensitivity in postmenopausal women and women,” Journal of the National Cancer Institute, Vol. the attenuating effect of added progestin,” Fertil Steril, 101, pp. 48–60, 2009. Vol. 60, pp. 664–667, 1993. [18] M. J. Borugian, S. B. Sheps, C. Kim-Sing, C. Van Patten, [31] G. K. Reeves, K. Pirie, V. Beral, J. Green, and E. Spencer, and J. D. Potter, “Insulin, macronutrient intake, and “Cancer incidence and mortality in relation to body mass physical activity: are potential indicators of insulin resis- index in the Million Women Study: Cohort study.” British tance associated with mortality from breast cancer?” Medical Journal, Vol. 335, pp. 1134, 2007. Cancer Epidemiol Biomarkers Prevention, Vol. 13, pp. [32] B. S. Connolly, C. Barnett, K. N. Vogt, J. Stone, and N. F. 1163–1172, 2004. Boyd, “A meta analysis of published literature on waist- [19] M. L. Garmendia, A. Pereira, M. E. Alvarado, and E. to-hip ratio and risk of breast cancer,” Nnutrition and Atalah, “Relation between insulin resistance and breast Cancer, Vol. 44, pp. 127–138, 2002. cancer among chilean women,” Annals of Epidemiology, [33] N. Kumar, A. Cantor, K. Allen, and C. Cox, “Android Vol. 17, pp. 403–409, 2007. obesity at diagnosis and breast cancer carcinoma [20] R. Kaaks, E. Lundin, S. Rinaldi, J. Manjer, and C. Biessy, survival,” Cancer, Vol. 88, pp. 2751–2757, 2000. “Prospective study of IGF-I, IGF-binding proteins, and [34] P. Pasanisi, F. Berrino, M. de Petris, E. Venturelli, and A. breast cancer risk, in northern and southern Sweden,” Mastroianni, “Metabolic syndrome as a pronostic factor Cancer Causes Control, Vol. 13, pp. 307–316, 2002. for breast cancer recurrences,” International Journal of [21] P. J. Mink, E. Shahar, W. D. Rosamond, A. J. Alberg, and Cancer, Vol. 119, pp. 236–238, 2006.

[35] B. A. Stoll, “Upper abdominal obesity, insulin resistance [45] C. R. Jonas, M. L. Mc Cullough, L. R. Teras, K. A. and breast cancer risk,” International Journal of Obesity Walker-Thurmond, and M. J. Thun, “Dietary glycemic and Related Metabolic Disorders, Vol. 26, pp. 747–753, index, glycemic load, and risk of incident breast cancer in 2002. postmenopausal women,” Cancer Epidemiol Biomarkers [36] A. S. Furberg, M. B. Veierod, T. Wilsgaard, L. Berstein, Prevention, Vol. 12, pp. 573–577, 2003. and I. Thune, “Serum high-density liporotein cholesterol, [46] S. Bowker, P. Veugelers, S. Majumdar, and J. Johnson, metabolic profile, and breast cancer risk,” Journal of the “Increased cancer-related mortality for patients with type National Cancer Institute, Vol. 96, No. 15, pp. 1152–1160, 2 diabetes who use sulfonylureas or insulin,” Diabetes 2004. Care, Vol. 29, pp. 254–258, 2006. [37] M. D. Holmes, W. Y. Chen, D. Feskanich, C. H. Kroenke, [47] S. Larsson, C. Mantzoros, and A. Wollk, “Diabetes mel- and G. A. Colditz, “Physical activity and survival after litus and risk of breast cancer: A meta analysis,” Interna- breast cancer diagnosis,” Journal of the American Medical tional Journal of Cancer, Vol. 121, pp. 856–862, 2007. Association, Vol. 293, pp. 2479– 2486, 2005. [48] P. Muty, T. Quattrin, B. J. B. Grant, V. Krogh, and A. [38] J. A. Ligibel, N. Campbell, A. Partridge, W. Y. Chen, and Micheli, “Fasting glucose is a risk factor for breast cancer: T. Salinardi, “Impact of a mixed strength and endurance a prospective study,” Cancer Epidemiol Biomarkers Pre- exercise intervention on insulin levels in breast cancer vention, Vol. 11, pp. 1361–1368, 2002. survivors,” Journal of Clinical Oncology, Vol. 26, pp. [49] E. Barrett-Connor, S. Slone, G. Greendale, D. Kritz- 907–912, 2008. Silverstein, and M. Espeland, “The Postmenopausal Es- [39] A. Pijpe, P. Manders, R. M. Brohet, J. M. Collée, and S. trogen/Progestin Interventions Study: Primary outcomes Verhoef, “Physical activity and the risk of breast cancer in in adherent women,” Maturitas, Vol. 27, pp. 261–274, BRCA 1-2 mutation carriers.” Breast Cancer Research, 1997. 2009. [50] A. Fournier, S. Mesrine, M. C. Boutron-Ruault, and F. [40] A. S. Fairey, K. S. Courneya, C. J. Field, G. J. Bell, L. W. Clavel-Chapelon, “Estrogen-progestogen menopausal hor- Jones, and J. R. Mackey, “Effect of exercise training on mone therapy and breast cancer: Does delay from meno- fasting insulin, insulin resistance, insulin-like growth fac- pause onset to treatment initiation influence risks?” Jour- tors, and insulin like growth factor binding proteins in nal of Clinical Oncology, 2009. post menopausal breast cancer survivors: A randomised [51] W. K. Hou, Y. X. Xu, T. Yu, L. Zhang, and W. W. Zhang, controlled trial,” Cancer Epidemiol Biomarkers and Pre- “Adipocytokines and breast cancer risk,” Chinese Medical vention, Vol. 12, pp. 721–727, 2005. Journal, Vol. 120, pp. 1592–1596, 2007. [41] V. Chajès, A. C. Thiébaut, M. Rotival, E. Gauthier, and [52] J. H. Kang, B. Y. Yu, and D. S. Youn, “Relationship of V. Maillard, “Association between serum trans-monoun- serum adiponectin and resistin levels with breast cancer saturated fatty acids and breast cancer risk in the E3N- risk,” Journal of Korean Medical Science, Vol. 22, pp. EPIC Study,” American Journal of Epidemiology, Vol. 117–121, 2007. 167, pp. 1312–1320, 2008. [53] S. Perrier, F. Caldefie-Chézet, and M. P. Vasson, “IL-1 [42] U. Risérius, “Trans fatty acids and insulin resistance,” family in breast cancer: Potential interplay with leptin and Atheroscler, Supplement Vol. 7, pp. 37–39, 2006. other adipocykine,” FEBS letter, Vol. 583, pp. 259–265, [43] B. A. Stoll, “Alcohol intake and late-stage promotion of 2009. breast cancer,” European Journal of Cancer, Vol. 35, pp. [54] K. Hancke, D. Grubeck, N. Hauser, R. Kreienberg, and J. 1653–1658, 1999. M. Weiss, “Adipocyte fatty acid-binding protein as a [44] L. S. Augustin, L. Dal Maso, C. La Vecchia, M. Parpinel, novel prognostic factor in obese breast cancer patients,” and E. Negri, “Dietary glycemic index and glycemic load, Breast Cancer Research and Treatment, Vol. 119, pp. and breast cancer risk: A case-control study,” Annals of 367–377, 2010. Oncology, Vol. 12, pp. 1533–1538, 2001.

9771833171009 01