TRANSLATIONAL RESEARCH IN CANCER: PRECLINICAL
PHARMACODYNAMICS AND CANCER EPIDEMIOLOGY
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy
in the Graduate School of The Ohio State University
By
Jia Ji, M.A.S.
Graduate Program in Pharmacy
The Ohio State University 2009
Dissertation Committee:
Dr. Jessie L.-S. Au, Advisor
Dr. M. Guillaume Wientjes
Dr. Dennis B. McKay
Copyright by
Jia Ji
2009
ABSTRACT
Translational research bridges preclinical and clinical research and shortens distance between these two areas in biomedical research. In preclinical study, our laboratory has found suramin sensitization at low and non-toxic dose and antagonism at toxic dose in multiple experimental models. The first part of this dissertation, Chapter 2 and 3, aims to evaluate cellular pharmacodynamics (PD) of biphasic effect of suramin and identify potential PD endpoint for future application in clinical practice. In Chapter 2, we first established an in vitro experimental model that illustrated suramin sensitization and antagonism to cisplatin, known as DNA-damaging drug. Addition of low dose suramin enhanced cellular response to cisplatin-induced DNA damage in three aspects, which are cell cycle arrest, cell death and senescence. Chapter 3 documented that persistence of γH2AX, a marker of DNA damage, was sustained by the addition of low dose suramin compared with cisplatin-alone treatment. No significant change of γH2AX kinetics was detected when suramin biphasic effect to taxanes, non-DNA-damaging drugs, was observed under both in vitro and in vivo settings. These preclinical discoveries not only lead us to treatment-dependant mechanism of suramin sensitization effect, but also indicate prospective clinical application of γH2AX as PD endpoint in anti-cancer therapy combined with suramin.
ii
The second part of this dissertation applied the principle of translational research for the purpose of studying and facilitating the application of research findings to the community. In cancer epidemiology, Chapter 4 investigated prevalence of serological response to human papillomavirus (HPV) 6, 11, 16, and 18, among women in China, as persistent HPV infection is a leading cause of cervical cancer. As the first report of HPV seroprevalence in China, we proposed necessity of primary prevention of HPV infection through application of HPV prophylactic vaccines, when most of Chinese women are not exposed. Chapter 5 focused on causes of breast cancer and ovarian cancer attributable to oral contraceptives use and reproductive factor change. Modest fraction of breast cancer and ovarian cancer is attributable to reproductive factor change, as insignificant percentage of cancer cases and deaths of breast cancer attributable to oral contraceptives use. These clinical findings prompt appropriate adoption of policies suitable and applicable to public health in China.
iii
DEDICATION
Dedicated to my grandmother, a strong woman, and my husband, whose encouragement and love is always with me
iv
ACKNOWLEDGEMENTS
I wish to express my sincere appreciation to my adviser, Dr. Jessie L.-S. Au, who has made this dissertation come true. Her scientific guidance, insightfulness and superior foresight have been inspiring me throughout my graduate study. It was her kind patience and persistent support that saved me from confusion and frustration during my most difficult times. I am always speechless on how to describe my gratefulness for her supervision, encouragement and strong faith on me, because no words can fully express my feeling. It has been my honor and privilege to receive such intact and extensive training on translational research from her. Her huge success in career and life will definitely influence and motivate me to pursue my dream.
My deep appreciation also goes to Dr. M. Guillaume Wientjes, for all his invaluable contribution to my overall scientific training and growth. His scientific enthusiasm, integrity and dedication are treasures I will be learning from. I am fortunate to have him as mentor and friend for life.
I truly thank Dr. Dennis B. McKay for being my committee member and all his intellectual advices and suggestions on my progress and dissertation. I genuinely appreciate his kindness and patience. I want to express my thankfulness to Dr. Duxin
Sun, who had guided me at the beginning of my graduate study and served as committee
v
member in my oral exam. I will remember his kindness, considerateness and
understanding and keep my deep appreciation to him from bottom of my heart.
A special recognition is extended to Dr. Myron Cohen and Dr. Jennifer S. Smith,
my mentors in the States, when I studied in Beijing as Fogarty scholar. It is their
dedication that made my last year productive. Sincere gratefulness goes to Dr. Youlin
Qiao, my mentor in China, for his invaluable scientific suggestions, discussions, and
input into my dissertation work. His amazing enthusiasm and devotion have greatly
inspired me. I appreciate his thoughtfulness and support. Also, everyone in his group
deserves my sincere gratitude for their help and care. I am blessed to work with them.
Everyone in the lab has played a special role in my training. I am really grateful to have such an enjoyable environment to finish the most important training in my career. I want to thank each of them for the sweet memorable days. In particular, my sincere thanks go to Dr. Yong Wei, Dr. Leijun Hu, Dr. Jie (Jack) Wang, Dr. Yan Xin and Dr. Bei
Yu, who walked me through the starting stage in this laboratory. They have earned my
undying gratitude for invaluable comments and encouragement. I would also like to
recognize the following individuals for their direct experimental contribution and
invaluable suggestions to my dissertation: Dr. Zancong Shen for the cooperation of
micro-autoradiography, Dr. Mingjie Liu, Dr. Ling (Lucy) Chen, Jianning Yang and Tong
Shen for the collaboration in pharmacodynamics study, Dr. Ji-hyun Chung for training on
western blotting and running samples, Dr. Ze Lu for providing tumor tissues, Dr. Ling
vi
Chen for the help of tissue sectioning and Xie (JJ) Zhe for some counting work. In
addition, I want to express my thankfulness to the following labmates for their friendship
and help: Dr. Xiao (Shelley) Hu, Dr. Greg Lyness, Dr. Dan Lu, Dr. Colin Walsh, Dr. Bin
Chen, Dr. Ho-lun Wong, Dr. Peng Guo, Jing Li, Yue Gao. My appreciation also goes to
Kathy Brooks, Carol Camm and Emily Noble for their patient help during my study.
My dear officemates, Jianning Yang and Jing Li, once again earned my special thankfulness for their friendship and encouragements during the last few years. Thanks to all my other friends for their support during the whole training. I will treasure those touching moments.
Last but not the least, my deepest appreciation goes to my grandmother, my husband, my mother, my father, my god-parents, my parents-in-law, my uncle and aunt.
Without their endless love and support, I would not be able to go through all the way here.
vii
VITA
August 22, 1979………………………………...Born – Beijing, P.R.China
July 2001….…………………………………….B.S., Pharmacy School of Pharmacy Peking University Health Science Center
August 2008….………………………………….M.A.S., Statistics The Ohio State University
April 2004 to present…………………...……….Graduate Research Associate College of Pharmacy The Ohio State University
July 2008 to June 2009………………………….Fogarty International Clinical Research Scholar Fogarty International Center National Institute of Health
PUBLICATIONS
Papers
1. Hao Cheng, Xianhua Cao, Ming Xian, Lanyan Fang, Tingwei Bill Cai, Jacqueline Jia Ji, Josefino B. Tunac, Duxin Sun, and Peng George Wang. Synthesis and Enzyme- Specific Activation of Carbohydrate-Geldanamycin Coujugates with Potent Anticancer Activity. Journal of Medicinal Chemistry, 48(2): 645-652, 2005.
2. Qin E’de*, He Xionglei*, Tian Wei*, Liu Yong*, Li Wei*, ……Ji Jia, ……Zhu Qingyu, and Yang Huanming. A Genome Sequence of Novel SARS-CoV Isolates: the Genotype, GD-Ins29, Leads to a Hypothesis of Viral Transmission in South China. Genomics, Proteomics & Bioinformatics, 1(2):101-107, 2003.
3. Xu Zuyuan*, Zhang Haiqing*, Tian Xiangjun*, Ji Jia* et al.. The R Protein of SARS- CoV: Analysis of Structure and Function Based on Four Complete Genome
viii
Sequences of Isolates BJ01-BJ04. Genomics, Proteomics & Bioinformatics, 1(2):155- 165, 2003.
4. Wang Jingqiang*, Ji Jia*, Ye Jia*, Zhao Xiaoqian* et al.. The Structure Analysis and Antigenicity Study of the N Protein of SARS-CoV. Genomics, Proteomics & Bioinformatics, 1(2):145-154, 2003.
5. Wu Qingfa*, Zhang Yilin*, Lv Hong*, Wang Jing*, ……Ji Jia, ……Yang Huanming. The E Protein Is a Multifunctional Membrane Protein of SARS-CoV. Genomics, Proteomics & Bioinformatics, 1(2):131-144, 2003.
6. Li Jingxiang*, Luo Chunqing*, Deng Yajun*, Han Yujun*, ……Ji Jia, ……YANG Huanming. The Structural Characterization and Antigenicity of the S Protein of SARS-CoV. Genomics, Proteomics & Bioinformatics, 1(2):108-117, 2003.
7. Wu Qingfa, Ji Jia, Dong Wei. Recent development in the study of Pharmacogenomics. Biotechnology, 12(2):39-41, 2002.
Presentations
1. Jia Ji*, He Wang*, Jennifer S. Smith, Mark Esser, Christine Velicer, Wen Chen, Shangying Hu, Robert G. Pretorius, Jerome L. Belinson, You-Lin Qiao. Population- based Seroprevalence of Human Papillomavirus in Chinese Women. The 25th International Papillomavirus Conference, Malmö, Sweden, May 8-14th, 2009.
2. Jia Ji, Yuli Chang, Jennifer S. Smith. Age-specific Prevalence of Human Papillomavirus DNA and Antibody: Systematic Review. The 25th International Papillomavirus Conference, Malmö, Sweden, May 8-14th, 2009.
3. Jennifer S. Smith , Jerome L. Belinson, Jia Ji, Shangying Hu, Wen Chen, Mark Esser, Frank J. Taddeo, Robert G. Pretorius, You-Lin Qiao. Population-based Human Papillomavirus 16/18/6/11 DNA and Seropositivity in Chinese Women. The 25th International Papillomavirus Conference, Malmö, Sweden, May 8-14th, 2009.
4. Shangying Hu, Jennifer S. Smith , Wen Chen, Jia Ji, Robert G. Pretorius, Mark Esser, Frank J. Taddeo, Jerome L. Belinson , You-Lin Qiao. HPV16/18 High in Precancer, Low in Normal Cytology in China. The 25th International Papillomavirus Conference, Malmö, Sweden, May 8-14th, 2009.
ix
5. Jia Ji, Yong Wei, Mingjie Liu, Zancong Shen, Tong Shen, M. Guillaume Wientjes, and Jessie L.-S. Au. γH2AX as pharmacodynamic marker in DNA-damaging treatment. College Research Day, College of Pharmacy, the Ohio State University, Columbus, OH, May 19th, 2008.
6. Hao Cheng, Xianhua Cao, Ming Xian, Lanyan Fang, Tingwei B. Cai, Jacqueline J. Ji, Peng G. Wang and Duxin Sun. Geldanamycin-carbohydrate prodrugs for enzyme- specific activation in cancer therapy. Proceedings of the American Association for Cancer Research Annual Meeting 98, Los Angeles, CA, 2007.
*These authors contributed equally to this work.
FIELDS OF STUDY
Major Field: Pharmacy
Studies in Pharmaceutics
x
TABLE OF CONTENTS
Abstract……………………………………………………………………………………ii
Dedication………………………………………………………………………………...iv
Acknowledgements………………………………………………………………………..v
Vita……………………………………………………………………………………...viii
List of Tables……………………………………………………………………………xvi
List of Figures…………………………………………………………………………..xvii
1. BACKGROUND INFORMATION ...... 1 1.1 Introduction...... 1 1.2 DNA damage and response...... 2 1.3 Activation of cell cycle checkpoints by DNA damage...... 3 1.3.1 DNA damage checkpoints ...... 3 1.3.2 G1 to S checkpoint (G1/S)...... 4 1.3.3 Intra-S phase checkpoint...... 4 1.3.4 G2 to M checkpoint (G2/M) ...... 5 1.3.5 M phase checkpoint (Spindle checkpoint)...... 5 1.4 Cellular responses to unrepaired DNA damages ...... 6 1.4.1 Apoptosis ...... 6 1.4.2 Senescence ...... 6 1.5 DNA damage induced by irradiation and chemotherapy...... 7 1.6 Repair of DNA damage caused by irradiation and chemotherapy ...... 9 1.6.1 Repair of DNA damage caused by irradiation...... 9 1.6.2 Repair of DNA damage caused by cisplatin...... 10 1.7 H2AX phosphorylation in DNA damage...... 11 1.7.1 H2AX phosphorylation as a DSB marker...... 11 1.7.2 Regulation of H2AX phosphorylation and dephosphorylation...... 11 1.7.3 Function of H2AX phosphorylation in DNA damage response...... 12 1.7.3.1 Apoptosis and γH2AX ...... 13 1.7.3.2 Senescence and γH2AX...... 13 1.8 Suramin and DNA damage response ...... 14
xi
1.8.1 Suramin...... 14 1.8.2 Extracellular and intracellular distribution of suramin ...... 14 1.8.3 Suramin and its possible targets in DNA damage response ...... 15 1.9 Human Papillomavirus (HPV) and cervical cancer...... 15 1.9.1 Cervical cancer...... 15 1.9.2 HPV in the etiology of cervical cancer...... 16 1.9.3 Other HPV-related diseases ...... 17 1.10 Markers of HPV exposure ...... 17 1.10.1 HPV DNA detection ...... 17 1.10.2 Serological response to HPV in human ...... 18 1.11 Prevalence of HPV exposure worldwide ...... 19 1.11.1 Prevalence of HPV DNA worldwide...... 19 1.11.2 Prevalence of serological response to HPV worldwide...... 19 1.12 Attributable cancers...... 20 1.12.1 Risk factors of cancer: genetic factors and environmental factors...... 20 1.12.2 Relative risks...... 20 1.12.3 Attributable fractions ...... 21 1.13 Attributable causes of breast cancer and ovarian cancer ...... 22 1.13.1 Risk factors of breast cancer and ovarian cancer...... 22 1.13.2 Reproductive factors ...... 23 1.13.3 Oral contraceptives ...... 24 1.14 Overview of the dissertation ...... 24
2. CHARACTERIZATION OF CISPLATIN-INDUCED TOXICITY AND ENHANCEMENT BY ADDITION OF SURAMIN...... 27 2.1 Introduction...... 27 2.2 Materials and methods ...... 28 2.2.1 Chemicals and supplies...... 28 2.2.2 Cell culture...... 29 2.2.3 Drug treatments...... 29 2.2.4 Micro-autoradiography ...... 29 2.2.5 Measurement of colony formation...... 30 2.2.6 Immunohistochemical staining of BrdU...... 31 2.2.7 Measurement of cell death...... 32 2.2.8 Measurement of apoptosis by Annexin V labeling...... 32 2.2.9 Measurement of senescence...... 33 2.2.10 Statistical analysis...... 34 2.3 Results...... 34 2.3.1 Intracellular location of suramin by micro-autoradiography...... 34
xii
2.3.2 Model establishment to study suramin chemosensitization under in vitro conditions ...... 34 2.3.3 Intra-S checkpoint by cisplatin and enhancement by addition of suramin ...... 35 2.3.4 Cell death by cisplatin and enhancement by addition of suramin...... 36 2.3.5 Apoptosis by cisplatin and change by addition of suramin ...... 36 2.3.6 Senescence by cisplatin and enhancement by addition of suramin...... 37 2.3.7 Delay in recovery from cisplatin treatment due to addition of low dose suramin...... 37 2.4 Discussion...... 38
3. EVALUATION OF EFFECT OF SURAMIN ON DNA DAMAGE INDUCED BY DNA-DAMAGING AND NON-DNA-DAMAGING AGENTS UNDER IN VITRO AND IN VIVO SETTINGS...... 53 3.1 Introduction...... 53 3.2 Materials and methods ...... 55 3.2.1 Chemicals and supplies...... 55 3.2.2 Cell culture...... 56 3.2.3 Drug treatments...... 56 3.2.4 Measurement of colony formation...... 56 3.2.5 Western blot analysis ...... 57 3.2.6 Immunohistochemical staining of γH2AX in cell lines...... 57 3.2.7 Immunochemical staining of M30 and γH2AX in in vivo tumor samples ...... 58 3.2.7.1 M30 staining ...... 59 3.2.7.2 γH2AX staining ...... 60 3.2.8 Statistical analysis...... 60 3.3 Results...... 60 3.3.1 Drug response of cisplatin on DSB...... 60 3.3.2 Single agent suramin did not enhance DSB...... 61 3.3.3 Addition of suramin to cisplatin altered DSB...... 61 3.3.4 Cytotoxicity of paclitaxel in FaDu cell lines ...... 62 3.3.5 Alteration of DNA damage by paclitaxel and effect of adding suramin in FaDu cells ...... 62 3.3.6 Suramin sensitization and antagonism to paclitaxel in SKOV3 cells ...... 63 3.3.7 Alteration of DNA damage by paclitaxel and effect of adding suramin in SKOV3 cells ...... 64
xiii
3.3.8 Apoptosis by docetaxel and enhancement by addition of suramin under in vivo setting ...... 65 3.3.9 Evaluation of DNA damage by docetaxel and alteration by addition of suramin under in vivo setting...... 65 3.3.10 Summary ...... 66 3.4 Discussion...... 66
4. SEROLOGIC RESPONSE TO HUMAN PAPILLOMAVIRUS TYPES 6, 11, 16 AND 18 IN CHINESE WOMEN...... 88 4.1 Introduction...... 88 4.2 Materials and methods ...... 89 4.2.1 Study population ...... 89 4.2.2 Sample collections ...... 90 4.2.3 HPV antibody detection by multiplex Luminex assay ...... 91 4.2.4 Statistical analysis...... 92 4.3 Results...... 92 4.3.1 Statistics of study subjects ...... 92 4.3.2 HPV seropositivity in sexually active women and virgins ...... 93 4.3.3 Age-specific sero-prevalence in Chinese women...... 93 4.3.4 Association between sociodemographic and sexual characteristics and seropositivity ...... 94 4.3.5 Seropositivity in women having normal cervix or with cervical lesions...... 95 4.3.6 Concordance between HPV DNA positivity and seropositivity...... 95 4.4 Discussion...... 97 4.5 Acknowledgements...... 104
5. ATTRIBUTABLE CAUSES OF BREAST CANCER AND OVARIAN CANCER TO REPRODUCTIVE FACTORS AND ORAL CONTRACEPTIVES IN CHINA ...... 116 5.1 Introduction...... 116 5.2 Materials and methods ...... 117 5.2.1 Data used for relative risk estimates of reproductive factors...... 117 5.2.2 Data used for exposure prevalence of reproductive factors...... 117 5.2.3 Data used for RR estimates of oral contraceptives ...... 119 5.2.4 Data used for exposure prevalence of oral contraceptives...... 120 5.2.5 Cancer incidence and mortality data...... 120 5.2.6 AF calculation...... 120 5.3 Results...... 121
xiv
5.3.1 Exposure prevalence of reproductive factors...... 121 5.3.2 Attributable fraction change of breast cancer and ovarian cancer to reproductive factor change ...... 121 5.3.3 Number of cancer cases and deaths attributable to reproductive factor change...... 122 5.3.4 Exposure prevalence of oral contraceptives ...... 122 5.3.5 Attributable fraction and number of cases and deaths of breast cancer attributable to oral contraceptives use...... 122 5.4 Discussion...... 123 5.5 Acknowledgements...... 127
6. PERSPECTIVES AND CONCLUSION...... 134 BIBLIOGRAPHY...... 137
xv
LIST OF TABLES
Table Page
2.1 Summary of pharmacodynamics of cisplatin with or without low and high dose suramin...... 39
4.1 Odds ratios (ORs) and 95% confidence intervals (CIs) for seropositivity according to sociodemographic and related variables ...... 107
4.2 Anti-HPV 16, 18 or 6/11/16/18 serological response and diagnostic approach to define cervical intraepithelial lesions...... 113
4.3 Comparison of HPV seropositivity and Hybrid Capture 2 result ...... 114
4.4 Comparison of HPV seropositivity and Linear Array result ...... 115
5.1 Number of Deaths and Cases of Breast and Ovarian Cancer in 2005 in China...... 128
5.2 Change in reproductive factors between 1982 and 2001 in China ...... 129
5.3 Change in AF between 1982 and 2001 in China ...... 130
5.4 Estimation of the number of breast and ovarian cancers cases and deaths in China in 2005 attributable to changes in reproductive risk factors between 1982 and 2001...... 131
5.5 Prevalence of current OC use in women 15–49 years old in China and attribute numbers of breast cancer (BC) cases and deaths...... 132
xvi
LIST OF FIGURES
Figure Page
2.1 Microautoradiographic images of PC3 histoculture treated with regular medium (control), 20 μg/ml 3H-suramin, and 200 μg/ml 3H-suramin...... 43
2.2 Cytotoxicity of one-hour cisplatin treatment in FaDu monolayer culture...... 44
2.3 Cell viability by suramin alone treatment (20 or 50 μM × 24 hr) on FaDu cell line.. 45
2.4 Biphasic effect of suramin at 5-100 μM on cisplatin treatment (30 μM ×1 h) in FaDu cell line...... 46
2.5 BrdU staining of FaDu cells at 2 and 4 days after cisplatin treatment with or without suramin...... 47
2.6 Cell detachment after cisplatin treatment or combined with suramin ...... 48
2.7 Apoptotic cell percentages after cisplatin treatment or combined with suramin...... 49
2.8 Effects of suramin on cisplatin-induced senescence...... 50
2.9 Growth kinetic curves after Cisplatin treatment with or without suramin in FaDu monolayer ...... 52
2.9 Growth kinetic curves after Cisplatin treatment with or without suramin in FaDu monolayer ...... 52
3.1 Suramin sensitization and antagonism to paclitaxel treatment (50 nM × 24 hrs) in FaDu cells ...... 70
3.2 Kinetic change of γH2AX expression (normalized to actin expression) after one-hour cisplatin treatment at different doses in FaDu monolayer ...... 71
xvii
3.3 Ratio of γH2AX/actin unchanged by suramin-alone treatment for 72 hr...... 73
3.4 Immunohistochemical stainning of γH2AX by suramin-alone treatment on day 7.... 74
3.5 Kinetic change of γH2AX expression (normalized to actin expression) after cisplatin treatment with or without suramin in FaDu monolayer...... 75
3.6 Immunohistochemical stainning of γH2AX by cisplatin treatment with or without suramin on day 7...... 76
3.7 Cytotoxicity of 24-hr paclitaxel treatment in FaDu monolayer culture...... 77
3.8 Kinetic change of γH2AX expression (normalized to actin expression) after paclitaxel treatment with or without suramin in FaDu monolayer (attached cells) ...... 78
3.9 Kinetic change of γH2AX expression (normalized to actin expression) after paclitaxel treatment with or without suramin in detached FaDu cells ...... 79
3.10 Cytotoxicity of 96-hr paclitaxel treatment in SKOV3 monolayer culture...... 80
3.11 Suramin sensitization and antagonism to paclitaxel treatment (20 nM × 96 hr) in SKOV3 cells ...... 82
3.12 Kinetic change of γH2AX expression (normalized to actin expression) after paclitaxel treatment (20 nM × 96 hr) with or without suramin in SKOV3 cells (both attached and detached cells)...... 83
3.13 In vivo apoptosis of docetaxel/suramin in A549 xenograft tumors after pretreatment of paclitaxel/carboplatin...... 84
3.14 Micrographs showing M30 immunohistochemical staining in A549 xenograft tumors after each treatment...... 85
3.15 In vivo γH2AX expression of docetaxel/suramin in A549 xenograft tumors after pretreatment of paclitaxel/carboplatin ...... 86
xviii
3.16 Micrographs showing γH2AX immunohistochemical staining in A549 xenograft tumors after each treatment...... 87
4.1 Seropositivity of anti-HPV 16, 18, 6, 11, 16 or 18, 6 or 11, and any of four types in sexually active women, virgins and overall women ...... 105
4.2 Age –specific HPV 6, 11, 16 and/or 18 seroprevalence rates in Chinese women.... 106
5.1 Age-stratified prevalence of current OC users in Chinese women (2001 data) ...... 133
xix
CHAPTER 1
BACKGROUND INFORMATION
1.1 INTRODUCTION
Significant efforts have been made to reduce the mortality rate, stabilize the
incidence rate, and improve the survival of cancer patients. Nevertheless, cancer still
accounts for more deaths than heart disease in persons under age 85 years in the United
States (1) and continues to be a major health problem worldwide. In China, there are 2.2
million new cases of cancer occurring in 2002, which is about one fifth of total incidence
in the world (2).
Translational research is an emerging discipline that bridges preclinical discovery
and clinical practice in a two-way direction. Scientific discoveries that begin from “the
bench” are in great need to be translated to clinical level, or the patient’s “bedside”.
Meanwhile, clinical observations offer the questions to be answered by preclinical
research. The work in this dissertation comprises preclinical and clinical studies on
cancer that were conducted in the United States and China.
The preclinical studies were focused on pharmacodynamics of suramin as a
chemosensitizer. Clinical effectiveness of current anti-cancer treatments is often
compromised by tumor resistance to these treatments that leads to disease recurrence.
1 Suramin, as a chemo- and radio-sensitizer, is studied in this dissertation for the
pharmacodynamics (PD) of its dose-dependant sensitization and antagonism effect, from the perspective of DNA damage and response in our preclinical experimental model. The clinical studies were focused on cancer epidemiology studies on the causes of breast cancer and ovarian cancer and the distribution of leading risk factors of cervical cancer in
China.
1.2 DNA DAMAGE AND RESPONSE
The primary structure of DNA is constantly subjected to alteration by endogenous
and exogenous DNA-damaging factors. These alterations can be simple base changes, or
more complex changes on the DNA backbone. There are mainly three types of DNA
damage: (a) DNA base damages including reduction and oxidization reaction on all four
types of DNA bases; (b) damages to the DNA backbone, e.g., single- and double-strand
DNA breaks; and (c) intra- and inter-strand links in DNA, or DNA-protein cross-links
(3).
DNA damage induces several cellular responses including checkpoint activation,
DNA repair, apoptosis pathways, and permanent growth arrest. Although these four response pathways have distinctive functions, they usually share the same set of molecular components to recognize and transduce the same signals. Therefore, activation of proteins involved in cell-cycle arrest also leads to induction of genes that participate in
DNA repair, apoptosis, and senescence. The pathways in each response are detailed as follows.
2
1.3 ACTIVATION OF CELL CYCLE CHECKPOINTS BY DNA DAMAGE
The cell division cycle consists of two components, mitosis where a single cell
divides into two daughter cells and interphase during which the cell prepares for mitosis.
The latter can be divided into three phases, G1 phase, S phase and G2 phase. Entry of a
cell into each phase is dependent on the proper progression and completion of the
previous one. Cell cycle checkpoints are to monitor and regulate the progress of the cell cycle, preventing cell cycle progression at specific points and allowing verification of necessary phase requirements and repair of DNA damage.
1.3.1 DNA damage checkpoints
Damages to DNA activate the four major cell cycle checkpoints (G1/S, intra-S phase, G2/M and M phase checkpoints), leading to slow down of cell cycle progression, thus allowing for time for DNA repair and recovery, and prevent DNA lesions from leading to harmful mutations. When DNA damage is not repairable, checkpoint
machineries can also initiate pathways leading to apoptosis or permanent cell-cycle
arrest. Checkpoints function to monitor genomic integrity, regulate cell proliferation, and
maintain genomic stability (4). Loss or attenuation of checkpoint function may increase
occurrence of gene mutations and chromosomal aberrations by reducing the efficiency of
DNA repair (5). Many types of cancer show mutations in components of signaling
pathways of cell cycle checkpoints.
A common pathway in the four checkpoints is the detection of DNA damage by
damage sensor proteins followed by transduction of signals to ATM (ataxia telangiectasia
mutated), ATR (ataxia telangiectasia mutated and Rad3-related protein), Chk1 and Chk2
3 kinases, which regulate effectors such as the Cdc25 phosphotases and p53 to inactivate cyclin-dependent kinases (Cdks).
1.3.2 G1 to S checkpoint (G1/S)
The G1/S cell cycle checkpoint controls the passage of cells from the first 'gap' phase, G1 phase, into the DNA synthesis phase, S phase. By inhibiting the initiation of
replication, it prevents cells from entering the S phase in the presence of DNA damage. If the DNA damage is double strand breaks, ATM is activated and phophorylates and activates its substrates including Chk2 and p53. Phophorylation of Chk2 and the consequential inactivation of Cdc25A rapidly initiate G1/S checkpoint while p53 activation maintains such arrest. The ATR-Chk1-Cdc25A pathway is alternatively activated when other kinds of DNA damage, such as cross-links, occur (3).
1.3.3 Intra-S phase checkpoint
The major mission of the cell division cycle is a faithful and complete duplication
of the genome followed by an equal partitioning of chromosomes to subsequent cell
generations. The precise replication of the genome is essential to cell survival. Therefore,
the intra-S phase checkpoint is particularly important in cell cycle regulation.
The activation of S-phase checkpoint induced by the generation/presence of DSB is manifested by a decreased rate of DNA synthesis. This protective mechanism delays the cells from completing the DNA replication, but not causing permanent arrest of cells at incompletely replication phase (6). The ATM-Chk2-Cdc25A-CDK2-Cdc45 axis emerges as a key mechanism of the transient intra-S-phase response. While rapid activation of ATM plays the early and dominant role in response to DSBs, ATR is locally
4 activated at later time to serve as a backup but necessary pathway through activating
Chk1 (7).
1.3.4 G2 to M checkpoint (G2/M)
Another important cell cycle checkpoint is G2/M checkpoint, which controls the
transition of cells from G2 to M phase. This checkpoint can prevent cells from undergoing mitosis in the presence of DNA damage. This provides an opportunity for
DNA repair and stopping the proliferation of damaged cells.
The ATM-Chk2-Cdc25/Wee1-Cdc2 and/or ATR-Chk1-Cdc25/Wee1-Cdc2
pathways are activated to arrest the cell in G2 by ionizing radiation or UV light,
respectively (3). As ATM can be activated rapidly in response to ionizing radiation, ATR primarily mediates the response to agents that induce cross-links, such as cisplatin.
However, they can serve as a backup function in response to the other stress. Both ATM
and ATR can phosphorylate either Chk1 or Chk2. Then they inhibit phosphatase Cdc25,
which blocks the activation of Cdc2 and then leads to arrest before entering M phase (8).
1.3.5 M phase checkpoint (Spindle checkpoint)
The checkpoint regulation in the M phase involves multiple steps in several subphases. The checkpoint proteins interact with kinetochores that mediate the binding of spindle microtubules to chromosomes in mitosis (9), and thereby monitor and ensure the fidelity of chromosome segregation in mitosis. Detection of failure of spindle attachment and improper spindle alignment results in arrest of cells in metaphase.
5 1.4 CELLULAR RESPONSES TO UNREPAIRED DNA DAMAGES
Failure to repair DNA damages can lead to apoptosis or senescence, as follows.
1.4.1 Apoptosis
Apoptosis, also called programmed cell death, is distinguished from other kinds of cell death by involving a series of biochemical events leading to specific cell
morphology and ultimate cell death. Therefore, it is an orderly process and controlled by
a group of cell signals. Cells with unrepaired DNA damage undergo apoptosis, through
the following mechanisms. Unrepaired DNA damage activates p53, which is phosphorylated by ATM, stimulates a sequence of reactions resulting in the cleavage of
Bid, or directly initiates the permeability transition of the mitochondrial membrane (10).
Several factors, including cytochrome C, are released into cytoplasm. Caspase-9 and the
rest of caspase cascade are activated to lead to apoptotic cell death.
DNA-damaging treatments can kill cancer cells by inducing and promoting apoptosis, which is one way to hamper tumor growth by anti-cancer treatments.
However, in recent years, many studies have shown that tumor resistance is, in part, possibly from failure to induction of apoptotic process, due to mutation of some of molecules involved in programmed cell death, such as p53.
1.4.2 Senescence
Senescence, also called irreversible proliferation arrest, was initially identified as a state associated with cell cycle arrest that occurs after cells have gone through a certain number of divisions in vitro. Once cells enter senescence, they cease to divide and
undergo a series of dramatic morphologic and metabolic changes, usually becoming
enlarged and flattened. Such viable but non-dividing state can be maintained for months
6 in senescent cells. Subsequent studies have shown that senescence can be activated in
response to various types of stress. Cellular senescence can be classified into two major
categories. Replicative senescence occurs after extended proliferation, and is triggered by
intrinsic mechanism such as telomere shortening. Stress-induced premature senescence
(or accelerated senescence) refers to rapid senescence triggered by extrinsic stress, such
as DNA damage, oxidative stress, and oncogene activation (11). Direct DNA damage by
radiation or DNA-damaging reagents can induce tumor cells to undergo senescence in
vitro and in vivo (12).
The decision of a cell to undergo apoptosis or senescence appears to depend on the type and the intensity of treatment and the cell type; higher levels of damage often
cause apoptotic death and lower levels often cause senescence (11). A recent body of
evidence suggests that induction of senescence, to certain extent, can be exploited as a
basis for cancer therapy. Senescence induction by chemotherapy results in a stable
disease rather than the regression of tumors (12), and represents a less aggressive
approach to control tumor progression.
1.5 DNA DAMAGE INDUCED BY IRRADIATION AND CHEMOTHERAPY
A number of endogenous and exogenous insults can cause damages to DNA. The
major type of DNA damage induced by ionizing radiation (IR) is the formation of DNA
DSBs. DSB is considered the most cytotoxic damage (13). Ionizing radiation causes
DSBs formation by targeting on DNA directly. DSB appearance has been believed to be
clinically significant lesions resulting from ionizing radiation.
7 Most chemotherapeutic agents induce DNA damage either directly or indirectly.
Intercalating agents (e.g. doxorubicin), topoisomerases inhibitors (e.g. irinotecan) and
alkylating agents (e.g. platinum compounds) are among the most well-known DNA-
damaging drugs used in the clinic.
Doxorubicin and irinotecan (CPT-11) are topoisomerase II and I inhibitors,
respectively. DSB is considered the most cytotoxic lesions induced by topoisomerase
inhibitors, due to stabilizing the cleavable complexes formed between enzymes and DNA
during replication. In DNA replicating cells, these complexes strike on the progressing
DNA replication fork and thus DNA in these complexes are converted into secondary
lesions consisting of DSBs (14; 15). Doxorubicin interacts with DNA through
intercalation and inhibition of DNA replication and inhibits topoisomerase II, an enzyme
involved in winding or unwinding both DNA strands during replication process. CPT-11
interferes with the function of topoisomerase I, an enzyme that is responsible for winding
or unwinding of single DNA strands.
Cisplatin is one of the most actively used drugs in clinical chemotherapy. It is a
neutral inorganic, water-soluble, coplanar complex. Its monoaquated active metabolite
reacts with cellular DNA to form interstrand and intrastrand crosslinks, which inhibits
DNA replication and RNA transcription and leads to DNA breaks and miscoding (16).
Although the primary lesions induced by cisplatin do not involve DSB, DSB can be produced from a fraction of single strand break (SSB) during DNA repair (17). Unlike those DNA-damaging agents that directly induce DSB, cisplatin-induced DSB occurs at a later time compared to irradiation (observed immediately) or other DNA damaging
agents such as DNA topoisomerase I and II inhibitors (observed within 2 hours) (17).
8 Emergence of DSB was also detected within the nuclear matrix after treatment of
mitomycin C, another potent DNA crosslinker (18).
Other types of anticancer drugs can trigger DSB as well. For example,
gemcitabine, a nucleotide analogue, is antimetabolites effective in cancer treatment
through replacing cytidine, stalling of replication fork and causing tumor growth arrest
and apoptosis. The nuclear foci of DNA damage were most evident in S-phase cells due
to stalled replication fork (19).
1.6 REPAIR OF DNA DAMAGE CAUSED BY IRRADIATION AND
CHEMOTHERAPY
Corresponding to various types of DNA damage, DNA repair can be divided into
five categories: direct repair (e.g. by photolyase), base excision repair (to remove single
base), nucleotide excision repair (to remove bulky DNA lesions), double-strand break
repair (by homologous recombination (HR) or nonhomologous end-joining (NHEJ)), and
repair of interstrand cross-links (by multiple mechanisms) (3). The latter two pathways
are prominent in DNA-damaging treatment and will be detailed below.
1.6.1 Repair of DNA damage caused by irradiation
For ionizing radiation-induced DSB, non-homologous end joining (NHEJ)
process accounts for most important repair pathway of DSB. In NHEJ, Two free DNA ends of little homology or no homology are joined and ligated. It can carry out nucleolytic, polymerization, and ligation operations on each strand independently (20).
This iterative processing nature of NHEJ is ideal for repair of DSB because it permits sequential action of the NHEJ enzymes on each DNA end and on each strand (20). The
9 core components of this pathway are the DNA-dependent protein kinase (DNA-PK),
Artemis, XRCC4 and DNA ligase IV (21; 22).
DNA-PK, a member of PI-3 kinase family, is composed of a catalytic subunit,
DNA-PKcs, and its regulatory subunits, Ku70 and Ku80. It is a molecular sensor for
DNA damage that magnifies the signal via phosphorylation of many downstream targets.
In the presence of DSB, Ku binds to the ends of broken DNA, followed by the recruitment of DNA-PKcs to constitute the active kinase complex (11). DNA-PK is then activated by DSB formation (13; 21; 22). One of its downstream targets is the minor histone H2A variant, H2AX.
1.6.2 Repair of DNA damage caused by cisplatin
It is not clear if a single protein or combinations of proteins are involved in sensing cisplatin-induced DNA damage (16). The major type of repair mechanism for cisplatin-induced cross-links is nucleotide excision repair (NER). Bulky adducts generated by cisplatin are removed from DNA through NER, in which single strand breaks (SSBs) are produced (23). NER repair pathway includes several steps: damage recognition, dual incisions bracketing the lesion to form oligomer, release of the excised oligomer, repair synthesis to fill in the resulting gap, and final ligation. Excision nuclease is the most important enzyme in this series of repair events (3). DSB converted from SSB can be repaired by NHEJ, the minor repair pathway served in repairing of cisplatin- induced DNA damage (17).
10 1.7 H2AX PHOSPHORYLATION IN DNA DAMAGE
1.7.1 H2AX phosphorylation as a DSB marker
Histones are basic nuclear proteins that are responsible for the nucleosome structure of the chromosome in eukaryotes. There are four core histones to form nucleosomes with DNA, called H2A, H2B, H3 and H4, respectively. H2AX, member X
of the histone H2A family, is systematically found and ubiquitously distributed
throughout the genome. The unique feature of H2AX in DNA damage and repair is that it
can be rapidly phosphorylated on Ser 139 of its carboxyl tail in the chromatin micro-
environment surrounding a DNA DSB (13; 24). Phosphorylated H2AX (gamma-H2AX,
γH2AX) is detected within 3 min after irradiation, rising to the plateau level in 10-30 min
(25). Therefore, γH2AX is a well-defined marker for DSB.
1.7.2 Regulation of H2AX phosphorylation and dephosphorylation
DNA damage induces histone H2AX phosphorylation by the phosphoinositide 3-
kinase like kinases (PI-3 kinases) such as ATM, ATR and DNA-PKcs. DNA-PKcs
phosphorylates H2AX in the NHEJ repair pathway, ATM or ATR are primarily
responsible for modifying H2AX in DNA damage checkpoint pathway, depending on the
type of damage cells encountered. Among these three kinases, ATM is believed to be the
major kinase to phosphorylate H2AX in normal cells (26).
Type 2 protein serine/threonine phosphatase (PP2A) is responsible for
dephosphorylation of γ-H2AX, with MDC1 as potential regulator to control the
dephosphorylation process. Dephosphorylation of γH2AX by PP2A is involved in
removal of γH2AX foci, which occurs during or after DNA repair. Once DNA damage is
removed, cells are released from checkpoint held-off and continue on cell cycle progress.
11 PP2A inhibition caused persistence of γH2AX foci and thus inefficient DNA repair, resulting in cell hypersensitivity to DNA damage (27). As a result, phosphorylation of
H2AX not only reflects the extent of DNA double strand break, but also indicates the efficiency of the repair process.
Therefore, the H2AX phosphorylation status represents a balance between the kinase activity and the phosphatase activity.
1.7.3 Function of H2AX phosphorylation in DNA damage response
H2AX plays an important role in DNA damage response. It is a major component of damage sensors (3). H2AX is involved in intra-S and G2/M checkpoints. It can be phosphorylated to γH2AX by ATM/ATR in response to DNA damage and forms DNA damage foci to strengthen DNA damage checkpoint activation to allow DNA repair.
In the repair of DSB which is activated by DNA-PK, γH2AX plays a role by being a binding site of repair proteins to damaged nuclear foci. It recruits and/or retains
DNA repair proteins such as BRCA1, MRE11/RAD50/NBS1 complex, MDC1 and
53BP1 (13). γH2AX also serves to help form irradiation induced foci (IRIF). Multiple studies have suggested a relationship between increased γH2AX persistence and unrepaired DNA damage and cell death (13).
Therefore, H2AX phosphorylation regulates both checkpoint proteins and repair factors in DNA damage response acting as a link between these two events. γH2AX is also found to be involved in the signaling pathways of two cellular fates, apoptosis and senescence.
12 1.7.3.1 Apoptosis and γH2AX
H2AX phosphorylation was previously thought to be a consequence of apoptotic
cell death in DNA damage response. Recently, it is found to contribute to apoptosis
through phosphorylation by JNK kinase (28). In addition to members of PI-3 kinase
family, ATM, ATR and DNA-PK, which can phosphorylate H2AX, a recent study found
that cJun NH2-terminal kinase (JNK), a member of MAP kinase family, is able to
phosphorylate H2AX in vitro upon exposure to UVA radiation. It showed that H2AX
phosphorylation is critical for DNA degradation and DNA ladder formation, but not for
the activation of caspase-3. Their results indicated that neither UVA-induced H2AX
phosphorylation nor caspase-3 activation alone is sufficient to induce apoptosis,
suggesting that JNK-H2AX pathway may cooperate with caspase-3 pathway to result in
cellular apoptosis (28).
Another study (29) showed that DSB formation, marked by H2AX
phosphorylation, may be an early event in the apoptotic pathway initiated by exposure to
DNA-damaging drugs. Rogakou et al. (30) also found that during apoptosis, γH2AX formation is induced by initiation of DNA fragmentation and is an early chromatin modification before the appearance of internucleosomal DNA fragments and the externalization of phosphatidylserine to the outer membrane leaflet. These findings support that H2AX phosphorylation precedes the initiation of apoptosis rather than being a consequence of apoptosis.
1.7.3.2 Senescence and γH2AX
Senescence is also associated with presence of γH2AX. In senescent cells, DSB can be formed during telomere shortening or from DNA-damaging agents (31). Intrinsic
13 stress that triggers occurrence of senescence can exert similar signal to those induced by
DSB from DNA damage stress, DNA damage foci are formed and then DNA damage
response pathway is switched on and mediated by the ATM/ATR (11). Senescence has
been viewed as permanently maintained DNA damage responses, which would involve
the presence of DNA damage foci containing γH2AX (32).
1.8 SURAMIN AND DNA DAMAGE RESPONSE
1.8.1 Suramin
Suramin, an organic polyanion with six sulfonic groups, is a polysulfonated
naphthylurea. Suramin is a drug with a long and rich history. Since the 1920s, it has been
used as an anti-parasitic drug, active in part because of its accumulation in the lysosomes
where the parasites reside. In the 1980s, suramin received attention as a possible
antiretroviral agent because it inhibits reverse transcriptase. Later, suramin has been
evaluated as an antineoplastic agent, where its mechanism of action was believed to be
inhibition of various growth factors, although many other pharmacologic actions were
also identified.
1.8.2 Extracellular and intracellular distribution of suramin
In the extracellular matrix, suramin can compete to interact with many growth
factors, presumably due to its structure similarity to glycosaminoglycans (33). Especially,
it binds to multiple fibroblast growth factors that are present in normal and malignant tissue (34), and binds on the cell surface with low affinity at multiple binding sites (35).
Structurally, suramin is a large polyanionic with very low pKa values. Hence, it is
not expected to readily diffuse across the cell membrane, which is a major mode of drug
14 transport. Many studies have shown that suramin can be internalized into the cell by a combination of the active processes of adsorptive endocytosis and pinocytosis.
Internalization is dependant on time of exposure and drug concentration (35). By
excytosis, suramin is effluxed from the cell, probably from endosomes or lysosomes.
1.8.3 Suramin and its possible targets in DNA damage response
Although there are very few studies in the actions of suramin on DNA damage
response, suramin can inhibit many proteins involved in γH2AX regulation, cell cycle
activation, DNA repair, apoptosis and senescence pathways. For example, suramin
inhibits DNA-PK and is a potent inhibitor of Cdc25A, the key mediator of intra-S phase
cell cycle checkpoint (36).
1.9 HUMAN PAPILLOMAVIRUS (HPV) AND CERVICAL CANCER
1.9.1 Cervical cancer
Cervical cancer is malignant tumor of the cervix uteri or cervical area. According
to GLOBOCAN 2002 (2), cervix cancer is the seventh most common cancer in the world
and the second most common cancer among women, with an estimated 493,000 new
cases and 274,000 deaths in the year 2002. Overall, it is much more common in
developing countries. There are relatively low age-standardized incidence rates of 6.8 per
100,000 and mortality rates of 3.8 per 100,000 in China (2). However, considering big
population base and limited prevalence data available, it is estimated to be around
100,000 new cases of cervical cancer occurring every year in China, accounting for about
one fifth of new cases in the world. Due to inadequate knowledge on cervical cancer (37)
15 and lack of country-wide screening program, cervical cancer posts a serious threat to
Chinese women.
Pathological diagnosis on cervical carcinoma precursor lesions made
classification of mild, moderate or severe dysplasia or carcinoma in situ (CIS). The term
of cervical intraepithelial neoplasia (CIN) was introduced later and CIN1, 2, 3 refer to
mild, moderate or severe dysplasia/CIS, respectively. Observation of squamous cell
abnormalities by cytologists in Papanicolaou (Pap) smear or liquid-based cytology was
defined in Bethesda system (38; 39). Low-grade Squamous Intraepithelial Lesion (LSIL)
corresponds to CIN1 and High-grade Squamous Intraepithelial Lesion (HSIL) includes
CIN2 and CIN3. Other than these two, atypical squamous cells of undetermined
significance (ASCUS) is one common category among mild abnormal results.
1.9.2 HPV in the etiology of cervical cancer
Zur Hausen H. proposed HPV infection might be a cause of cervical cancer and
soon found abundant prevalence of HPV 16 DNA in cervical cancer biopsy tissues (40).
Since then, numerous studies confirmed persistent infection of specific HPV types plays
an etiological role in cervical cancer development. To date, more than 130 types of HPV
have been identified. Thirteen of them are carcinogenic to human beings and listed as high-risk types (41). HPV 16 and 18, two most common oncogenic types, are responsible for 70% of cervical cancer worldwide.
HPV are double-stranded circular DNA viruses. Many HPV types that infect
genital tract are primarily through sexual transmission (42). There is a strong association
between sexual activity and HPV infection. Other transmission pathways might exist due
to findings of genital HPV infection in virgins (43).
16 1.9.3 Other HPV-related diseases
Condylomata acuminata (genital warts) is a common sexually transmitted disease, caused by infection of some types of HPV (44). HPV 6 and 11 are identified to account for 90% of cases. These two types are not implicated in cervical cancer development and thus listed as low risk types. Several HPV types are associated with cancer of the vulva, vagina, penis, anus, oral cavity, and oropharynx, and in less association with cancer of the larynx and periungual skin (41; 44).
1.10 MARKERS OF HPV EXPOSURE
Because of the etiological role of HPV in cervical cancer, there is a strong quantitative relationship of HPV exposure and risk of cervical cancer. There are two major markers of HPV exposure, HPV DNA testing and serological response measurement.
1.10.1 HPV DNA detection
Persistent HPV infection can induce progression to precancer in the cervix.
However, 90% of HPV infections are transient and viral clearance occurs in an average of 12 to 24 months (45). Because most HPV infections can not be defined by visual or microscopy inspection, HPV DNA detection becomes a reliable option in measuring
HPV exposure. Therefore, HPV DNA detection is found to be correlated well with number of recent sexual partners (46) as a marker of recent exposure to HPV.
Compared to cytology diagnosis, HPV DNA testing is more sensitive but less specific. Hybrid Capture 2 (HC2) by Qiagen Corp. has been approved by FDA as the only HPV testing used in cervical cancer screening. It is an in vitro nucleic acid
17 hybridization assay with signal amplification using microplate chemiluminescence. It
measures total positivity of 13 high-risk types and that of 5 low-risk types, separately,
though not detecting specific types (47). On the other hand, HPV genotyping, mostly
using PCR methods, can identify positivity of each specific type, giving higher sensitivity than HC2.
1.10.2 Serological response to HPV in human
Although most HPV infections can self-clear, antibody against HPV is stable over
time once seroconversion occurs (46). Measurement of HPV antibody in serum is also
favored in areas conservative to sexual culture, considering sampling difficulties because
many unmarried women in China are unwilling to undergo gynecologic examinations for
the collection of exfoliated cells. In contrast to HPV DNA infection as a marker of recent
HPV exposure, HPV antibody response can be used as a reliable indicator of past or
cumulative exposure to HPV.
There are a number of antibody tests developed to detect serum antibody against
HPV with fair sensitivity and specificity. They include in vivo neutralization assays, in
vitro pseudoneutralization assays, competitive radioimmunoassays (cRIAs), and enzyme-
linked immunosorbent assays (ELISA). Enzyme-linked immunosorbent assay (ELISA) is
most often used, due to technical convenience and comparability to cRIAs. Recently,
multiplexd Luminex assay emerges as to simultaneously quantitate neutralizing
antibodies to multiple HPV types, which is suitable for population-based studies on sero-
prevalence or monitoring vaccine responses in clinical trials (48).
18 1.11 PREVALENCE OF HPV EXPOSURE WORLDWIDE
1.11.1 Prevalence of HPV DNA worldwide
With abundant evidence of HPV as a cause of cervical cancer and other related
disease, investigation on prevalence of HPV infection is in great need for the purpose of cancer control and prevention. Global data showed that at any given time point, about
10% of women carry HPV infections though they do not show any symptoms (49). This suggests HPV is one of the most common sexually transmitted infections. HPV 16 is the most common type and HPV 18 the second in women with normal cervix or having cervical lesions worldwide (50). Smith et al. reviewed age-specific prevalence of HPV infection in the world (51). Across regions, rates of HPV infection are high in younger women and rapidly decline in middle-aged women. There is inconsistency in HPV DNA prevalence in older women, with a decrease or plateau or even an increase of HPV infection. Sexual behavioral difference may be most likely cause for worldwide variation of HPV infection with age or in regions (49; 51).
1.11.2 Prevalence of serological response to HPV worldwide
Compared with thousands of studies detecting HPV infection in populations, there
is much less survey on sero-prevalence of HPV worldwide. Nevertheless, immuno-
response to HPV is an important marker of life-time exposure. Knowledge on HPV
antibody prevalence is essential for the future implementation of HPV vaccines in
cervical cancer prevention. Since most sero-prevalence studies focused on a few specific
types of HPV, anti-HPV 16 remains the most common type in most studies but anti-HPV
6 or 11 became more prevalent if tested. Given delayed seroconversion and repeated
infection needed to stimulate immunological response, there is a peak prevalence of anti-
19 HPV in middle-aged women worldwide, often followed by a decrease or plateau of seroprevalence in older women (52).
1.12 ATTRIBUTABLE CANCERS
1.12.1 Risk factors of cancer: genetic factors and environmental factors
Risk factors in cancer development include genetic susceptibility, unhealthy
lifestyle, and environmental influence. The first one represents intrinsic factor and the
latter two stand for extrinsic evidence. Although cancer is a disease characterized by
multiple genetic alterations, only a minority of all cancers are attributable to known
genetic susceptibility factors. In this dissertation, lifestyle and environmental factors are
combined as environmental factors or non-genetic factors (53), as opposed to genetic
factors. The proportion of cancers attributable to environmental factors is unknown in
many countries, except in some developed countries (54; 55). The weight of different risk
factors in the burden of cancer in a given population is critical for better understanding of
the relative importance of risk factors and for prioritization of public health efforts. Since
one-fifth of new cancer cases worldwide occurs in China every year (2), attributable
causes of cancers in China is in need of investigation to guide efforts in cancer control
and prevention.
1.12.2 Relative risks
In statistics and mathematical epidemiology, relative risk (RR) is the risk of an
event (or of developing a disease) relative to exposure. Relative risk is a ratio of the
probability of the event occurring in the exposed group versus a non-exposed group (56).
It should be differentiated from odds ratio, another important term in epidemiology. Odds
20 ratio is the ratio of the risk in the exposed divided by the risk in the unexposed. In a circumstance of small probability, relative risk is close to odds ratio in value. In medical research, the odds ratio is favored for case-control studies and retrospective studies.
Relative risk is used in randomized controlled trials and cohort studies.
1.12.3 Attributable fractions
The proportion of cancer in the total population that can be attributed to a risk factor is denoted the attributable fraction (AF) (57), and is expressed as a percentage
(%). This concept is used to estimate attributable causes of the burden of cancer. For cancer risk factors that can be avoided or completely suppressed, at least in theory, the most straightforward way to estimate the attributable fraction is to calculate the fraction of all cases (exposed and unexposed) that would not have occurred if exposure had not occurred (58). Hence in this configuration, the alternative (counterfactual) scenario to current exposure is the total absence of exposure. For cancer risk factors that cannot be completely avoided or suppressed, another approach consists of estimating the avoidable fraction of cancer that is the fraction of cancer that would not occur if an alternative
(counterfactual) scenario of attainable exposure level or exposure intensity was considered (59).
The AF can be calculated as a function of the relative risk (RR) of cancer associated with exposure to a risk factor and the prevalence of exposure (“P”) of a population to that risk factor. This method was originally described by Levin (60):
P *()RR −1 AF = [ P *()RR −1 ]+1
21 Prevalence of exposure data may only be available as continuous variables, but
not in percentages. For these exposures, a mean relative risk associated with each unit
increase in exposure can be derived from relative risks associated with exposure
categories (e.g., the increase in cancer risk for one unit of alcohol). For these exposures
for which only the average level is known and a RR per unit of exposure increase is
available, the following formula can be used:
AF = [exp(ln(RR)*Exposure Level)-1]/[exp(ln(RR)*Exposure Level)]
Using AF and cancer incidence and mortality data, number of cases and deaths of
cancer attributable to specific factor can be calculated. However, there should be a lag
time between at which time prevalence of specific factor was drawn and cancer statistics was obtained. The length of latency period may vary, depending on progress of influence
by specific factor on cancers.
1.13 ATTRIBUTABLE CAUSES OF BREAST CANCER AND OVARIAN
CANCER
1.13.1 Risk factors of breast cancer and ovarian cancer
A number of factors are indicated in the etiology of breast cancer. Mostly studied
factors having genetic influence are BRCA1 and BRCA2 mutation. Heritage factors, such
as family, race and age, are considered important risk factors to breast cancer (61).
Environmental factors are mostly hormone-related, including reproductive factors,
obesity, hormone therapy, oral contraceptives (OC), and oophorectomy or hysterectomy
history (62; 63).
22 Several factors are recognized to be involved in ovarian cancer occurrence. They consist of ovulation, gonadotropic and steroid hormones, germ cell depletion, oncogenes and tumor suppressor genes, growth factors, cytokines, and environmental agents. Family history, some reproductive factors, hormone-replacement therapy (HRT) is evidenced of a higher risk to breast cancer. Lifestyle factors like cigarette smoking and some environmental agents may increase the risk, but more confirming data are needed (64).
1.13.2 Reproductive factors
Several reproductive factors in the lifetime of a woman have been studied for their association with breast cancer. The list of these factors includes, but is not limited to, parity, age at menarche, age at menopause, age at first birth, age spacing of pregnancies, breast feeding, abortion history, and infertility (65). Some of these factors are identified in cancer development as either protective factor or risk factor. Others play a complicated role due to complex effects from hormonal production and regulation or interaction with other reproductive factors.
Risk factor of ovarian cancer involves reproductive component as well.
Reproductive factors such as age at menopause and infertility tend to increase the risk of ovarian cancer, whereas pregnancy, tubal ligation, and hysterectomy are protective against ovarian cancer (64). The protective effect of pregnancy might be from regulation of certain hormones associated with multiparity (64).
The latency period in AF study does not apply to reproductive factors, for which recent exposure is more important to determine cancer risk.
23 1.13.3 Oral contraceptives
Use of oral contraceptives (OC), containing similar chemicals to hormone replacement therapy (HRT), contribute to modest risk of breast cancer (66), but it has clearly been shown to reduce the risk of ovarian cancer (64). Because current OC use is associated with breast cancer and such risk would disappear after cessation of OC use
(67). No lag time was considered for OC use in AF analysis.
1.14 OVERVIEW OF THE DISSERTATION
The rest of this dissertation is divided into five chapters. Each chapter begins with a brief introduction of background information and the purpose of the study, followed by a description of the methods and materials used. Experimental or analytical findings are summarized in the results section, followed by more detailed discussion regarding the significance and implications of the findings in the discussion section. Tables and figures are attached at the end of each chapter, while references are included at the end of the dissertation.
Chapter 2 studied the establishment of an in vitro model to study suramin chemosensitization effect to cisplatin treatment in FaDu monolayer. Using this model, we evaluated cisplatin-induced cytotoxicity from three major aspects in DNA damage and response, which are cell cycle arrest, apoptosis and senescence. Alteration of cellular responses to cisplatin treatment by addition of low and high dose suramin was presented.
At selected dose of cisplatin, recovery delay on cell growth by addition of low dose suramin was demonstrated by clonogenic assay. The possibility of intracellular location of suramin was also explored by micro-autoradiography.
24 Chapter 3 evaluated the intensity of DNA damage associated with different doses
of cisplatin through monitoring expression of γH2AX, the marker of DSB. All above
evidence suggested that low-dose suramin enhanced and sustained DNA damage induced
by cisplatin by observation on kinetics change of γH2AX. All of these experiments were done with the model established in Chapter 2 under in vitro setting. Then we evaluated the intensity of DNA damage associated with different doses of taxanes, considered as
non-DNA-damaging drugs. Using a previously established model that suramin caused
sensitization and antagonism to paclitaxel treatment, by monitoring change of γH2AX with time, there was no DNA damage change caused by addition of suramin under in vitro setting. Using A549 tumor tissues treated by docetaxel, DNA damage induced by docetaxel and alteration by addition of suramin was also insignificant under in vivo setting. Combined with the findings in Chapter 2, enhancement and retention of DNA damage induced by addition of low dose suramin is dependant on the type of chemotherapeutics.
Chapter 4 describes the clinical research experience in cancer epidemiology. Due to concern on increasing exposure to HPV in China, serological response to HPV 6, 11,
16 and 18 was investigated among 4,211 Chinese women. Age-specific prevalence of antibody response to these four common HPV types was reported. Association between sociodemographic and sexual characteristics and seropositivity was presented. The findings led to the proposal of preventing HPV infection in China.
Chapter 5 focused on attributable causes of breast cancer and ovarian cancer in
China. Five representative reproductive factors were chosen to study attributable causes of these two female cancers in China to women’s reproduction history. Exposure data in
25 1982 and 2001 on these five factors was provided. Using cancer incidence and mortality
data in 2005, attributable fraction of breast cancer and ovarian cancer to reproductive factor change was calculated and number of cancer cases and deaths attributable was reported, respectively. Prevalence of current oral contraceptives use was listed in each age group and attributable fraction of breast cancer to oral contraceptives use was
estimated among all women.
The final chapter summarizes the conclusions and contributions of this
dissertation research and discusses future investigations.
26
CHAPTER 2
CHARACTERIZATION OF CISPLATIN-INDUCED TOXICITY AND
ENHANCEMENT BY ADDITION OF SURAMIN
2.1 INTRODUCTION
Cisplatin, a platinum chemical in anti-cancer treatment, induces its cytotoxic activity through producing DNA cross-links and adducts. As the first discovered metal complexes with anticancer activity, cisplatin has been successfully used in clinic for several types of solid tumors. However, its effectiveness has been compromised by toxic side effects and tumor resistance that often leads to tumor re-occurrence (68). On the other hand, chemosensitization effect of suramin at low or non-toxic dose of 10-50 µM was found in multiple experimental models, including various chemotherapeutic drugs in different xenograft tumor models or in tumor cell lines (69-71). Noncytotoxic suramin enhanced chemosensitivity of tumor cells and did not increase toxicity from chemotherapy, achieving significant benefit in combinational therapy. This sensitization effect was also observed in vivo for combination with radiation therapy, whose mechanism of cytotoxic effect is believed to be direct damage on DNA in tumor cells
(unpublished results by Dr. Yan Xin).
27 Biochemical mechanism of action of cisplatin involves DNA binding, interference with transcription/replication, and production of cisplatin-DNA adducts. Cell death is
ultimately induced when cisplatin-DNA adducts are not efficiently processed by cell
repairing machinery (72). Cellular response to DNA damage provoked by cisplain is
manifested in three major aspects: cell cycle checkpoint activation, apoptosis pathway, and permanent growth arrest (senescence). Meanwhile, suramin may have multiple targets implicated in biochemical pathways of DNA damage response.
In this chapter, the intracellular location of suramin was evaluated first. Then we aimed to establish an in intro model to demonstrate suramin chemosensitization to cisplatin treatment. We also observed the pharmacodynamics (PD) of cisplatin-induced cellular responses and the alterations due to addition of suramin.
2.2 MATERIALS AND METHODS
2.2.1 Chemicals and supplies
Cis-Diammineplatinum(II) dichloride (cisplatin), suramin sodium salt, crystal violet, 5-Bromo-2’-deoxyuridine (BrdU) were purchased from Sigma-Aldrich Co. (St.
Louis, MO), cell culture supplies (i.e. RPMI Medium 1640, L-glutamine, non-essential
amino acids, antimitotic-antimycotic, fetal bovine serum or FBS, Trypsin-EDTA) and
Phosphate Buffered Saline (PBS) from Invitrogen Corporation (GIBCO) (Carlsbad, CA);
[3H(G)]-suramin sodium (specific activity 12.4 Ci/mmol) from Moravek Biochemicals
Inc. (Brea, CA), Kodak autoradiography Emulsion Type NTB, Kodak developer D-19
and fixer from Eastman Kodak Company (Rochester, NY), methyl green nuclear
counterstain from Vector Laboratories Inc. (Burlingame, CA); Mouse anti-BrdU antibody
28 from Dako North America, Inc. (Carpinteria, CA), LSAB kit from Dako North America,
Inc. (Carpinteria, CA), liquid DAB kit from BioGenex (San Ramon, CA); Annexin V
Fluorescein Conjugate from Invitrogen Corporation (Molecular Probes) (Carlsbad, CA);
X-gal from EMD Biosciences, Inc. (San Diego, CA), and senescence β-Galactosidase
staining kit from Cell Signaling Technology, Inc. (Danvers, MA). All chemicals and
reagents were used as received.
2.2.2 Cell culture
FaDu cells (American Type Culture Collection, Rockville, MD) were maintained
as monolayer cultures at 37°C in a humidified atmosphere containing 5% CO2 in RPMI
1640 medium supplemented with 9% fetal bovine serum, 2 mM L-glutamine, 0.1 mM
non-essential amino acids, 1% antimitotic-antimycotic (100×). For experiments, cells were harvested from subconfluent cultures using trypsin and resuspended in fresh
medium before plating. Culture medium was replaced every two days.
2.2.3 Drug treatments
Cisplatin stock solutions (5 mM) prepared in sterile water were stored in the dark
at room temperature. Suramin stock solutions in sterile water at 20 mM were stored at
4ºC. Drug-containing medium was replaced every two days for long-term experiments.
2.2.4 Micro-autoradiography
The procedures of autoradiography was detailed somewhere else (73). PC3
histocultures were treated with RPMI 1640 medium containing [3H]-labeled suramin of
20 μg/ml (about 14 μM) or 200 μg/ml (about 140 μM) or control (suramin-free medium)
for 96 hr. The histoculture pieces were then frozen at -20ºC. Tissue was cryo-sectioned
and mounted onto slides that were pre-coated with emulsion (NTB nuclear emulsion,
29 Kodak) and placed in the dark. The slides were developed with Kodak developer and
fixer in a dark room and counter-stained with methyl green. The stained slides were
scanned at low magnification (1000 ×) using a Zeiss Axiovert 35 microscope (Carl Zeiss,
Thornwood, NY).
2.2.5 Measurement of colony formation
Colony-forming chemosensitivity assay, or clonogenic assay, could reflect the
contribution of all forms of drug-induced cytotoxicity including mitotic catastrophe,
apoptosis, senescence, and permanent growth arrest, by quantifying clonogenic survival
(74).
Cells were seeded in 48-well microtiter plates (16,000 cells in 0.2 ml per well)
and were allowed to attach to the surface by growing in drug-free medium for 20 to 24 hr.
Suramin-free or suramin-containing medium at doubled desired concentration in 0.2 ml
was added to each well and cells were incubated for another 24 hr. Then medium was
removed and cells were treated with control (drug-free medium), cisplatin, or
combination treatment of cisplatin and suramin, in 0.5 ml for 1 hr. Control or drug-
containing medium was instantly removed after treatment and cells were washed by drug- free medium in 1 ml for three times with 10 min for each wash. Cells were then incubated in 0.5 ml of suramin-free or suramin containing medium for specific period.
At each chosen time point after treatment, medium was removed and cells were trypsinized in 0.1 ml for about 5 min. Observed under microscope, when all cells were fully detached from the bottom, they were resuspended with 0.9 ml of drug-free medium completely to prevent any cell cluster or aggregation. Between 100 to 60,000 cells were seeded in 10-cm petri-dish and were allowed to attach to the bottom surface and form
30 colonies by growing in drug-free medium for 11 to 12 days. The seeding amount was
decided by cell counting and estimated cytotoxicity calculation, to make 100 to 200 formed colonies in each dish. Drug-free medium was replaced every six days to allow the treated cells to attach to the bottom completely, recover and form colonies in an undisruptive manner. Before each step, microscopic observation of cells status was performed.
When most of cell colonies reached the size of 30 or 40 cells, medium were removed and cells were washed with PBS for one time and stained with filtered ethanol- dissolved 0.4% crystal violet for 10 to 20 min. The dish was placed under running tap water to wash away the stain. Afterwards, dishes were dried and colonies of greater than
20 cells were counted. The clonogenicity in the untreated control ranged from 60% to
110%, with unknown causes of this big variation.
2.2.6 Immunohistochemical staining of BrdU
At 2 or 4 days after cisplatin treatment as indicated in 2.2.4, cells were labeled by
BrdU (50 μM) for 1 hr and then washed three times with cold PBS. After washing, cells were instantly fixed in cold methanol and stored at -20ºC for later staining. Anti-BrdU antibody was used to stain BrdU-labled cells. The staining was continued using LSAB and liquid DAB kits, and counter-stained with hematoxylin.
BrdU, an analogue of thymidine, is incorporated within cell nuclei and has been used to study cell cycle progression (75). Weak immunonistochemical BrdU staining reflects cells stalled in the replication fork. When counting cells, the stained cells were classified into weak (light brown, cells not fully covered by brown color) and strong
(dark brown, cells fully covered by brown color) stained cells. On average, we counted
31 722 ± 281, 396 ± 88, 432 ± 86, 412 ± 56 cells per replicate in the control, cisplatin, cisplatin/low dose suramin and cisplatin/high dose suramin groups on day 2, and 900 ±
365, 361 ± 121, 419 ± 140, 352 ± 103 cells per replicate in the control, cisplatin, cisplatin and low dose suramin and cisplatin and high dose suramin groups on day 4, respectively.
2.2.7 Measurement of cell death
Cell detachment was used to measure cell death induced by drug treatment. Cells were seeded in 6-well microtiter plates with 120,000 cells in 2 ml per well. The control wells were seeded at a lower density (60,000 cells for 1-day control, 40,000 for 2-day control, 12,000 for 4-day control, 1,200 for 7-day control) to accommodate faster cell growth, compared with treated wells. At 1, 2, 4 and 7 days after treatment, medium containing the detached cells was centrifuged and the remaining attached cells were harvested by trypsinization and centrifuged. After centrifugation at 2000 rpm for 10 min, the supernatant was discarded and the remaining cells were resuspended in PBS of 100 μl to 1 ml and the cell number was counted under microscope. Both cell density and cell volume were recorded to calculate total detached or attached cells in each well. The numbers of detached cells on progressive days were added to show the cumulative detachment over time. For example, data on day 4 represented the sum of detached cells on day 2 plus that on day 4. Similarly, data on day 7 equaled to the sum of day 2 plus day
4 plus day 7.
2.2.8 Measurement of apoptosis by Annexin V labeling
In the early phase of apoptosis, the membrane phospholipid phosphatidylserine is translocated from the inner to the outer leaflet of the plasma membrane, which can be recognized by Annexin V with high affinity (76). Propidium iodide (PI), which binds to
32 nucleic acids, can only penetrate the plasma membrane when membrane integrity is
breached which occurs in the later stages of apoptosis or in necrosis. Therefore, Annexin
V and PI labeling in flow cytometry indicates the percentages of early and late apoptosis
cells.
Cells were harvested by trypsinization, washed once with cold PBS, and then
resuspended with binding buffer with about 1 × 106 cells per ml. Five μl of Annexin V conjugate was added to each 100 μl of cell suspension. Cells were incubated for 15 min at room temperature. One μl propidium iodide (PI) was then added to each sample, followed by 400 μl of binding buffer. Each sample was mixed gently and kept on ice for immediate flow cytometry detection. We used Annexin V-labeled cells to represent apoptosis cells in either early or late phase.
2.2.9 Measurement of senescence
Senescence-associated β-galactosidase (SA-β-gal) marker was detected by a colorimetric assay using X-gal as a substrate at pH 6.0. At each time point after cisplatin- containing treatment, cells were washed once with PBS and fixed with fixative solution for 10 to 15 min. Then after two-time washing, cells were incubated with X-gal- containing staining solution at 37 ºC for overnight staining. Then cells were checked for blue color and overlaid with 70% glycerol for long-term storage at 4ºC. We counted at least 500 cells per replicate per time point. The seeding density of first and second experiment was 6×104 and 2.5×104 per well in 12-well plates, respectively. It explained
difference on percentage of senescent cells in magnitude due to positive correlation
between percentage of senescent cells and seeding density founding in pilot studies. Also
the first experiment used commercial kit and the second one self-prepared agents.
33 2.2.10 Statistical analysis
Statistical significance of the differences between two groups was assessed by the two-tailed Student’s t test. Differences were considered significant when p<0.05.
2.3 RESULTS
2.3.1 Intracellular location of suramin by micro-autoradiography
The micrographs presented in Figure 2.1 showed that suramin distributed
relatively evenly in both extracellular and intracellular space. The amount of radiation was dose dependant and was found in both nuclear and extra-nuclear areas. This raises the possibility of suramin acting on extracellular and intracellular components.
2.3.2 Model establishment to study suramin chemosensitization under in
vitro conditions
The inhibition curve on cell growth of FaDu cells by cisplatin treatment was measured by clonogenic assay and shown in Figure 2.2. Cells were treated with cisplatin
0.1, 0.3, 1, 2, 3, 5, 10, 20, 30, 50, 100 μM for 1 hr, and then seeded to form colonies right after treatment. Clonogenicity was calculated for each concentration, compared with control (no treatment). Cytotoxicity by cisplatin was observed from 1 μM and when cisplatin reached 50 μM, maximum cell killing effects were obtained. The cisplatin concentration (30 μM) used to study suramin sensitization is close to maximum cytotoxicity.
Single suramin treatment at 20 and 50 μM for 24 hr did not affect cell viability
validated by clonogenic assay (Figure 2.3). Biphasic effect of suramin on cisplatin
treatment in FaDu cell line was observed through clonogenicity measurements (shown in
34 Figure 2.4). Clonogenicity was about 0.7% by one-hour single cisplatin treatment at 30
μM. Combination treatment of cisplatin and suramin was used in our model as suramin was added 24 hr before cisplatin treatment and continue to treat cells throughout until 48 hr after stopping cisplatin treatment, which means suramin was both pre- and post-treated to cells. Addition of suramin at 5 to 20 μM greatly enhanced chemosensitivity of FaDu cells to cisplatin treatment by significantly decreasing clonogenicity. Maximum sensitization effect was observed at 20 μM of suramin added, which was about half of clonogenicity achieved by cisplatin treatment alone. However, pre- and post-treatment of suramin at 30 to 100 μM for the same time length, when combined with same cisplatin treatment, produced no effect or antagonistic effect as shown by no change or increased clonogenicity. We chose suramin at 20 and 50 μM to be combined with cisplatin treatment, namely low dose and high dose suramin, and established as an in vitro model used to study suramin sensitization and antagonism for most following studies. In combination treatments in section 2.3.3 through section 2.3.7, suramin was added 24 hr before cisplatin treatment and maintained in medium until specific time points of measurement after cisplatin treatment.
2.3.3 Intra-S checkpoint by cisplatin and enhancement by addition of
suramin
Based on treatment strategy established in section 2.3.2, we first investigated cell cycle arrest by cisplatin treatment and its alteration by addition of suramin. Cisplatin arrests cells at intra-S checkpoint (77). Uptake status of BrdU by cells can be used to check on progress of intra-S arrest, because BrdU is an analogue of thymidine. Cells stalled in the replication machinery showed incomplete or weak staining of BrdU,
35 whereas cells in normal S-phase progression have complete or strong staining (75). We
found that cisplatin treatment significantly enhanced the ratio of weak-to-strong BrdU
staining (Figure 2.5), which confirms intra-S phase progress was slowed down by
cisplatin. Suramin-alone treatment did not change S-phase progression by keeping the
same weak-to-strong ratio as control group (data not shown). Combinations of cisplatin
and low dose suramin gave higher weak-to-strong BrdU staining ratios compared to
cisplatin-alone treatment, at day 2 and day 4 after cisplatin treatment. In contrast,
combination of cisplatin and high dose suramin gave lower weak-to-strong ratio of BrdU
staining compared with cisplatin-alone treatment. This data suggests that addition of low
dose suramin sustained and enhanced the extent of arrest in intra-S checkpoints, whereas
high dose suramin lacked such effect.
2.3.4 Cell death by cisplatin and enhancement by addition of suramin
FaDu cells were treated with cisplatin with or without suramin as indicated in
section 2.3.2. Both attached and detached cells for each group were counted at day 1, 2,
4, and 7 after cisplatin treatment. The percentage of detached cells by cisplatin-alone
treatment was increasing from 1 to 7 days after treatment, with around 40% of detached
cells at day 7 (Figure 2.6). Cell viability as measured by clonogenicity was not affected by single agent suramin (Figure 2.3). The percentage of detached cells showed that cisplatin combining low dose suramin yielded more detached cells than cisplatin-alone or cisplatin combining high dose suramin at later time points, but not at early time points.
2.3.5 Apoptosis by cisplatin and change by addition of suramin
Annexin V labeling in flow cytometry was used to track the percentages of apoptotic cells (76). The result in Figure 2.7 showed that at 1, 2 and 4 days after cisplatin
36 treatment, addition of low dose suramin did not increase the percentage of apoptotic cells,
compared to cisplatin alone treatments. Addition of high dose suramin yielded a bit more
apoptotic cells at day 1 and 2, albeit not significantly.
2.3.6 Senescence by cisplatin and enhancement by addition of suramin
The changes of percentage of senescent cells among the three groups from day 1
to 7 after treatment were shown in Figure 2.8. An additional experiment that followed the kinetics change of senescent cells to 7 and 9 days after cisplatin treatment was also shown in Figure 2.8. Percentage of senescent cells was gradually increasing after cisplatin treatment and reached the highest point at day 6, followed by a decline at day 7 and day 9. There were no significant differences between cisplatin with and without suramin at early times after treatment, i.e., days 1 to 4. Addition of high dose suramin to cisplatin significantly decreased the fraction of senescent cells from day 5 through day 9, whereas addition of low dose suramin did not affect the fraction of senescent cells at all time points.
2.3.7 Delay in recovery from cisplatin treatment due to addition of low dose
suramin
Figure 2.9 showed the effect of suramin on the kinetics of cell growth after cisplatin treatment. Cell growth was measured by clonogenic assay. After cisplatin
treatment, cells began to recover slowly in the first two days with more rapid recovery
rate at the later days. Addition of low dose suramin delayed the recovery significantly
such that no recovery was observed at least for 7 days. In contrast, addition of high dose suramin showed more rapid recoveries in 2 of 3 experiments.
37 2.4 DISCUSSION
There are a number of in vitro assays to measure chemotherapeutic sensitivity in
tumor cells (78). A good assay is able to predict in vivo activity of anti-neoplastic drugs.
Cytotoxicity was initially evaluated by assays that measure changes of certain proteins or
enzymes or a specific process associated with cellular metabolism. Due to poor
predictability or limited application, these assays were improved or replaced by more
effective assay. Clonogenic assay was then introduced and became a prominent assay
with superior ability to predict chemo-sensitivity testing (79). This assay has been used in
broad applications, such as preclinical screening of new agents, cytogenetic analysis of
human tumor specimens, and the identification of growth factors and hormones for
different tumor types. Therefore, we chose this assay to evaluate tumor sensitivity
exclusively in our model. For consistency, inhibition curve by cisplatin in FaDu cells was
also measured by clonogenic assay. However, there are some limitations by clonogenic
assay (80). They include technical difficulties, disruption of normal cell-to-cell
interaction, and cell kill measured in a narrow log range. Quantitative difference between
experiments in our study may be due to plating efficiency, intrinsic cell characteristics, extrinsic environmental change etc.
When we established in vitro model to study suramin chemosensitization, it showed schedule-dependency in FaDu cells. We observed that addition of suramin with
24-hr pre-treatment and 48-hr post-treatment to cisplatin treatment (15 μM × 2.5 hr) had maximized chemosensitization effect, compared with either addition of pre- or post- treatment of suramin (unpublished results by Dr. Yong Wei). It indicates that suramin
38 must be present before and after cisplatin treatment to obtain maximized sensitization.
This treatment schedule was applied in our model thereafter.
We attempted to characterize the intracellular distribution of suramin. The micro-
autoradiography results suggested that suramin may be distributed in the intracellular or
extracellular space, which renders the possibility that suramin may interact with factors in
both spaces. One limiting factor is the high background. This may be improved by
following the strict technical requirements as described by Stumpf et al. (81).
Table 2.1 summarizes the quantitative measurements of several
pharmacodynamics endpoints of cisplatin effects by addition of suramin from section
2.3.3 to section 2.3.8.
cisplatin and low dose cisplatin and high dose suramin versus cisplatin suramin versus cisplatin treatment treatment Clonogenicity decrease (p<0.05) increase (p<0.05) trend of slight increase on trend of slight decrease on Intra-S arrest (BrdU staining) day 2 and 4 day 2 and 4 Cell detachment increase on day 7 (p<0.05) unchanged trend of slight increase on Apoptosis (Annexin V) Unchanged for 4 days day 1 and 2 trend of slight increase on decrease on day 5, 6, 7 Senescence day 7 and 9 and 9 (p<0.05) Cell recovery (growth kinetics) decrease (p<0.05) ??
Table 2.1 Summary of pharmacodynamics of cisplatin with or without low and high dose
suramin.
39 Addition of low dose suramin to cisplatin appeared to enhance cell cycle arrest at
early time and senescence at late time, while significantly increasing cell detachment,
reducing cell growth and delaying recovery from cisplatin treatment. Significantly
decreased clonogenicity by addition of low dose suramin can not rule out delayed
detachment after plating in clonogenic assay, due to significant cell detachment at late time by addition of low dose suramin. In contrast, significantly increased clonogenicity by addition of high dose suramin may be due to less of these cellular responses to cisplatin with less or equal cell-cycle arrested cells and significantly deceased senescent cells. Two experiments plus a pilot experiment showed inconsistent effects of addition of high dose suramin on cell recovery. Whether addition of high dose suramin accelerated or maintained cell recovery rate of cisplatin treatment was questioned as marked. More experiments would be conducted to investigate this critical finding.
Cisplatin is known to cause cell arrest at G1/S, intra-S, and G2/M checkpoints
(77). Addition of suramin sustained and enhanced the extent of arrest in intra-S checkpoints as shown in section 2.3.3 by BrdU uptake measurement. Cell cycle analysis using flow cytometry also observed that addition of low dose suramin to cisplatin treatment resulted in intra-S phase arrest at 1 and 2 day after treatment, showing a slightly slower rate to progress in the S phase than cisplatin alone treatment (unpublished results by Dr. Mingjie Liu). Cells arrested in intra-S phase may undergo incomplete DNA synthesis due to the presence of damaged DNA. Nocodazole, through depolymerization of microtubules, causes arrest of cells in metaphase and is used to study M phase entry
(82). In a separate study, in cells treated with nocodazole (5 μM × 1.5 hr) given at 2 days after cisplatin treatment, addition of low dose suramin reduced the fraction of M phase
40 cells, suggesting that low dose suramin assisted cisplatin-induced G2/M checkpoint arrest
(unpublished results by Dr. Yong Wei). This additional evidence further confirmed our observation that cellular response to cisplatin-induced DNA damage was enhanced by addition of low dose suramin regarding to cell cycle arrest.
More detached cells were induced by combination treatment of cisplatin and low dose suramin at later time points. Pilot experiments showed that there were very few colonies formed in detached cells by any of three treatments, indicating most detached cells lost clonogenicity (data not shown). This observation is consistent with the chemosensitization effect found in section 2.3.2 and suggests the lower clonogenicity caused by addition of low dose suramin to cisplatin might have resulted from changes associated with cell detachment.
Senescent cells represent cells arrested on proliferation permanently or irreversibly (11). Senescent cells modulate cancer progression (83) and lead to a less aggressive disease (12). Our in vivo study showed in mice bearing FaDu xenograft tumors, combination of cisplatin and low dose suramin produced more cases with stable diseases than cisplatin alone treatment and prolonged survival (unpublished results by
Tong Shen). The agreement between in vitro and in vivo observation suggested senescence as a potential mechanism of suramin sensitization effect.
These cellular pharmacodynamics effects correlate with the biphasic character of suramin effect to cisplatin. The collective data indicate that cytotoxic cellular responses induced by cisplatin treatment were generally enhanced and sustained by addition of low dose suramin featuring significantly increased cell detachment and delayed cell recovery.
Addition of high dose suramin significantly decreased senescent cells and possibly
41 accelerated cell recovery. Cell recovery appeared to be one critical PD effect of suramin chemosensitization. The biphasic observation on PD characteristics partially explained suramin sensitization and antagonism at different doses and shed light on the potential mechanism behind this intriguing effect.
42 Control Low dose suramin
High dose suramin
Figure 2.1 Microautoradiographic images of PC3 histoculture treated with regular
medium (control), 20 μg/ml 3H-suramin, and 200 μg/ml 3H-suramin
All pictures were taken under 1000× magnification. Sections of histoculture treated by low dose suramin were exposed for 11 days and those by high-dose suramin exposed for 4 days. All sections were counter-stained with methyl green after development and fixation.
43 120
100
80
60
40
20 Clonogenicity (% of control) of (% Clonogenicity 0 0.1 1 10 100 Cisplatin Concentration (uM)
Figure 2.2 Cytotoxicity of one-hour cisplatin treatment in FaDu monolayer culture. The details of experiment were described in section 2.3.2. Data are mean + SD (n=3 replicates
in 1 experiment). Some SDs are smaller than the symbols. The clonogenicity in the
untreated control was 82.3+10.0%.
44 150
100
50 Clonogenicity (% of control) of (% Clonogenicity 0 02050 Suramin Concentration (uM)
Figure 2.3 Cell viability by suramin alone treatment (20 or 50 μM × 24 hr) on FaDu cell line. There was no significant difference compared to untreated control. Data are mean +
SD (n=2 experiments with 3 replicates/experiment). The bar represents one SD. The clonogenicity in the untreated control was 65.6+10.3%.
45
Cisplatin Cisplatin (30 μM ×1 h) combined with suramin only 6 *
4
2 * * * * * 0 Clonogenicity (% of control) of (% Clonogenicity 0 5 10 20 30 50 100 Suramin Concentration (uM)
Figure 2.4 Biphasic effect of suramin at 5-100 μM on cisplatin treatment (30 μM ×1 h) in
FaDu cell line. The details of experiment were described in section 2.3.2. * indicates
P<0.05 compared to cisplatin alone treatment (by student’s t test). Data are mean + SD
(n=3 replicates in 1 experiment). The bar represents one SD. The clonogenicity in the untreated control was 90.6+7.2%.
46 A.
40
30 § * Cis30+S20 Cis30 20 *
staining Cis30+S50 § Cis30+S20 10 * # Cis30 control Cis30+S50
Ratio of weak:strong BrdU BrdU weak:strong of Ratio control 0 241 Days after Cisplatin treatment
B. Weak Strong
Figure 2.5 A. BrdU staining of FaDu cells at 2 and 4 days after cisplatin treatment with or without suramin.
Cis30: cisplatin treatment 30 μM × 1 h. Cis30+S20: cisplatin and suramin at 20 μM. Cis30+S50: cisplatin and suramin at 50 μM. * indicates P<0.05 compared to untreated control (by student’s t test). In combination treatments, suramin was added 24 hr before cisplatin treatment and maintained until day of measurement. # indicates P=0.05 compared to untreated control (by student’s t test). § indicates P=0.10 compared to untreated control (by unpaired student’s t test). There was no significant difference between cisplatin alone treatment and combined with suramin treatment (P=0.57 and 0.42 for 2 and 4 days, respectively. by ANOVA test). Data are mean + SD (n=3 replicates in 1 experiment). The bar represents one SD. B. BrdU staining of FaDu monolayer showing weak and strong staining. 47 80 Cis30 Cis30 + S20 * 60 Cis30 + S50
40 #
% detached cells 20
0 02468 Days after cisplatin treatment
Figure 2.6 Cell detachment after cisplatin treatment or combined with suramin. Cis30: cisplatin treatment 30 μM × 1 h. Cis30+S20: cisplatin and suramin at 20 μM. Cis30+S50: cisplatin and suramin at 50 μM. In combination treatments, suramin was added 24 hr before cisplatin treatment and maintained until day of measurement. * indicates P<0.05 compared to cisplatin alone treatment (by student’s t test). # indicates P=0.07. Data are mean + SD (n=3 replicates in 1 experiment). Some SDs are smaller than the symbols.
48 20 Control Cis30 16 Cis30+S20 Cis30+S50 12
8
4 % Annexin V-positive cells V-positive % Annexin 0 124 Days after cisplatin treatment
Figure 2.7 Apoptotic cell percentages after cisplatin treatment or combined with suramin.
Cis30: cisplatin treatment 30 μM × 1 h. Cis30+S20: cisplatin and suramin at 20 μM.
Cis30+S50: cisplatin and suramin at 50 μM. In combination treatments, suramin was added 24 hr before cisplatin treatment and maintained until day of measurement. There was no significant difference between untreated control and cisplatin alone treatment or between cisplatin alone treatment and combined with suramin treatment. Data are mean +
SD (n=5 experiments with 6 replicates in total at day 1 and 2, n=2 experiments with 3 replicates in total at day 4). The bar represents one SD.
49 A. Experiment 1
50 Cis30 Cis30 + S20 40 Cis30 + S50 30
* 20 * *
% Senescent cells 10
0 02468 Days after cisplatin treatment
Continued
Figure 2.8 Effects of suramin on cisplatin-induced senescence. Two experiments were performed, one for 7 days (Experiment 1) and one for 9 days (Experiment 2). Cis30: cisplatin treatment 30 μM × 1 h. Cis30+S20: cisplatin and suramin at 20 μM. Cis30+S50: cisplatin and suramin at 50 μM. In combination treatments, suramin was added 24 hr before cisplatin treatment and maintained until day of measurement.
A. Experiment 1: Kinetics of senescent cell percentage after cisplatin treatment or combined with suramin from day 1 to day 7. * indicates P<0.05 compared to cisplatin alone treatment (by student’s t test). Data are mean + SD (n=3 replicates in 1 experiment). Some SDs are smaller than the symbols. B. Experiment 2: Senescent cell percentage after cisplatin treatment or combined with suramin on day 7 and 9. * indicates P<0.05 compared to cisplatin alone treatment (by student’s t test). Data are mean + SD (n=3 replicates in 1 experiment). The bar represents one SD. C. Microscopic image of senescent cell staining at day 7 after cisplatin treatment.
50 Figure 2.8 Continued
B. Experiment 2
15 Cis30 12 Cis30+S20 Cis30+S50 9
6
% Senescent cells 3 * * 0 79 Days after cisplatin treatment
C.
Cisplatin 30 μM Cisplatin 30 μM + Suramin 20 μM
51 Experiment 1 Experiment 2
8 1.5 Cis30 Cis30 Cis30+S20 Cis30 + S20 1.2 Cis30+S50 6 Cis30 + S50 0.9 4 0.6 # 0.3 2
# control) of (% Clonogenicity Clonogenicity (% of control) of (% Clonogenicity * 0.0 # * 0 0246802468 Days after cisplatin treatment Days after cisplatin treatment
Experiment 3
24 Cis30 Cis30+S20 18 Cis30+S50 *
12
6 *
Clonogenicity (% of control) of (% Clonogenicity * 0 * 02468 Days after cisplatin treatment
Figure 2.9 Growth kinetic curves after Cisplatin treatment with or without suramin in
FaDu monolayer. Results from three experiments were shown. Cis30: cisplatin treatment
30 μM × 1 h. Cis30+S20: cisplatin and suramin at 20 μM. Cis30+S50: cisplatin and
suramin at 50 μM. In combination treatments, suramin was added 24 hr before cisplatin treatment and maintained until day of measurement. * indicates P<0.05 compared to
cisplatin alone treatment (by student’s t test). # indicates P=0.06. Data are mean + SD
(n=3 replicates/experiment). Some SDs are smaller than the symbols. The clonogenicities
in the untreated control were 111.5+19.8%, 58.9+8.6% and 107.3%+10.4%, respectively
in three experiments. 52
CHAPTER 3
EVALUATION OF EFFECT OF SURAMIN ON DNA DAMAGE INDUCED BY
DNA-DAMAGING AND NON-DNA-DAMAGING AGENTS UNDER IN VITRO
AND IN VIVO SETTINGS
3.1 INTRODUCTION
Suramin has been found by our laboratory to sensitize tumor cells to chemotherapy and radiotherapy at low and non-cytotoxic concentrations of 10-50 µM
(84), but antagonize at high and cytotoxic dose. The mechanisms that contribute to the biphasic effects remain largely unknown. An in vitro experimental model was established to study the phamacodynamics (PD) of this biphasic effect of non-cytotoxic suramin in
Chapter 2. The results showed that cellular response to DNA damage induced by cisplatin treatment is enhanced by addition of low dose suramin, but not affected by high dose suramin. Nevertheless, single suramin treatment did not affect cell survival, indicating
sensitization and antagonism occurs under the mechanism of suramin action on the
biochemical pathways influenced by cisplatin.
The major type of DNA damage induced by cisplatin is to form DNA crosslinks
and interrupt DNA replication and RNA transcription (16). Single strand breaks (SSBs)
produced through repair of cisplatin-induced DNA damage were converted to double
53 strand breaks (DSBs) (17). In the clinical treatment of cisplatin, the most concerned fact
has been the development of cisplatin resistance by tumors. This might be due to the
reduction in cisplatin accumulation inside cancer cells because of barriers across the cell
membrane, the faster repair of cisplatin adducts, the modulation of apoptotic pathways in
various cells, the loss of p53 and other mechanisms (85). FaDu cells used in our
experimental model has mutated p53 (86) and defective apoptotic pathways. Also, this in vitro model allows us to study cytotoxicity of drugs with the least barriers around the cells. On the other hand, addition of low dose suramin is found to enhance and sustain
DNA damage response activated by cisplatin treatment, in terms of three cellular reactions (cell cycle arrest, apoptosis and senescence). Suramin might be located intracellularly to target on specific factors involved in biochemical pathways of DNA damage response. Significant sensitization effect of low dose suramin was also observed when combined with other DNA-damaging treatment such as ionizing radiation therapy
(unpublished results by Dr. Yan Xin). This led to the hypothesis that non-cytotoxic suramin may enhance and sustain DNA damage induced by DNA-damaging treatment, e.g. cisplatin treatment.
DSB is the major cytotoxic DNA damage by ionizing radiation, cisplatin and many other DNA damaging treatments. Phosphorylation of histone protein, H2AX, or noted as γH2AX, is a bio-marker of DSB because it consistently occurs after DSB formation (87). Appearance of γH2AX plays a role in cellular response to DNA damage due to its involvement in cell cycle checkpoints activation, apoptosis and senescence triggering. Therefore, in the following study, we used γH2AX as marker of DNA damage to be evaluated after drug treatments.
54 Non-DNA-damaging treatments include antimetabolites (such as gemcitabine)
and mitotic disrupters (such as paclitaxel). Paclitaxel is one of the most active cancer chemotherapeutics. It causes cell death through promoting the polymerisation of tubulin and disrupting the normal microtubule dynamics required for M phase progression (88).
Using clonogenic assay, suramin sensitization and antagonism to paclitaxel treatment (50
μM × 24 hr) was observed in FaDu cells (Figure 3.1), which is similar to the model we established in chapter 2. Although non-DNA-damaging agents are not expected to cause
extension of DSB, DSB may occur as an independent PD endpoint, as an indirect consequence of drug effects.
In this chapter, we evaluated the PD of DSB by a prototype DNA damaging agent cisplatin and by prototype taxanes.
3.2 MATERIALS AND METHODS
3.2.1 Chemicals and supplies
In addition to chemicals and supplies as stated in Chapter 2, section 2.2.1, paclitaxel was purchased from Yunnan Hande Bio-Tech Company (Houston, TX), cell scraper from Thermo Fisher Scientific Inc. (Waltham, MA), western blotting equipments from Bio-Rad Laboratories, Inc. (Hercules, CA), protease inhibitor cocktail from Sigma-
Aldrich Co. (St. Louis, MO), BCA protein assay kit from Pierce Biotechnology, Inc.
(Rockford, IL), phospho-histone H2A.X (Ser139) antibody from Cell Signaling
Technology, Inc. (Danvers, MA), anti-Histone H2A.X from Upstate USA, Inc.
(Charlottesville, VA) or from Cell Signaling Technology, Inc. (Danvers, MA), Blotto non-fat dry milk and Actin (c-2) HRP mouse monoclonal antibody from Santa Cruz
55 Biotechnology Inc. (Santa Cruz, CA), beta-actin antibody, anti-mouse or anti-rabbit IgG
HRP-linked Antibody, biotinylated protein ladder detection pack and prestained protein marker from Cell Signaling Technology, Inc. (Danvers, MA), ECL plus western blotting detection system from GE Healthcare (Piscataway, NJ), GE Pure Nitrocellulose Transfer
Membrane from GE Osmonics Labstore (Minnetonka, MN), CL-XPosure Film from
Pierce Biotechnology, Inc. (Rockford, IL), photographic developer and fixer from OSU stores (Columbus, OH); albumin from bovine serum (BSA) and hydrogen peroxide 30% solution from Sigma-Aldrich Co. (St. Louis, MO), normal goat serum from Vector
Laboratories Inc. (Burlingame, CA), M30 CytoDEATH antibody from Roche Applied
Science (Indianapolis, IN), anti-phospho-H2AX (ser139) from Upstate USA, Inc.
(Charlottesville, VA) or Cell Signaling Technology, Inc. (Danvers, MA). All chemicals and reagents were used as received.
3.2.2 Cell culture
FaDu and SKOV3 cells were maintained as mentioned in Chapter 2, section 2.2.2.
3.2.3 Drug treatments
Drug treatments were given as mentioned in Chapter 2, section 2.2.3. Paclitaxel stock solutions were prepared in ethanol at 100 μM, stored in dark at -20ºC. Aliquots of paclitaxel stock solution were added to culture medium so that the final concentration of ethanol was less than 0.1%, which does not affect paclitaxel effects. Control group to paclitaxel treatment contained 0.05% ethanol.
3.2.4 Measurement of colony formation
Details on clonogenic assay was described in Chapter 2, section 2.2.5.
56 3.2.5 Western blot analysis
The procedure for Western blot analysis of γH2AX were as described elsewhere
(89). Briefly, cells cultured in dishes or plates were collected with a cell scraper and centrifuged at 5000 rpm for 10 min at 4ºC. Cell pellets were harvested and added lysis buffer of equal amount to extract protein. After centriguation at 14,000 rpm for 20 min at
4ºC, cell lysates were measured by BSA assay to quantitate protein concentration in each sample. Samples with equal amount of proteins were loaded on a SDS-PAGE gel and run at 70V for 2 hr, and then transferred onto nitrocellulose membranes overnight.
Nitrocellulose membrances were blocked in 5% milk in Tris-buffered saline containing
0.2% Tween 20 (TBST) at room temperature for 1 hr and then washed twice with the same buffer. The membrane was incubated with primary antibody (1:1000 dilution for
γH2AX) for 3 hr at room temperature and then washed 3 times with TBST. The secondary antibody, anti-rabbit IgG HRP-linked antibody, was applied at 1:2000 dilution in 5% milk in TBST for 1 hr at room temperature and then washed five times with TBST.
The bands were developed using ECL Western blotting detection reagents and analyzed using ImageJ software (version 1.37, National Institute of Health). The procedures for western blot analysis of H2AX and Actin were similar to γH2AX. The primary antibody was at 1:4000 dilution and anti-mouse IgG HRP-linked antibody was at 1:2000 dilution.
3.2.6 Immunohistochemical staining of γH2AX in cell lines
At 7 days after cisplatin or suramin treatment as indicated in 2.3.2, cells were trypsinized and centrifuged at 2,000 rpm for 10 min at 4ºC and then washed once with cold PBS. Cell pellets were instantly fixed in cold methanol by slowly dropping methanol into vortexing cells and stored at -20ºC for later staining. One or two drops of cell pellets
57 containing appropriate amount of cells were dropped onto glass slides and dried for
staining. The cells were rinsed with phosphate-buffered saline (PBS) once and blocked in
diluted hydrogen peroxide for 10 min and washed twice with PBS. The cells were
blocked in 5% goat serum in PBS for 1 hr at room temperature. Then anti-γH2AX antibody was used at 1:1000 dilution in PBS for 1 hr at room temperature. The staining was continued using LSAB and liquid DAB kits, and the cells were finally counter- stained with diluted hematoxylin.
3.2.7 Immunochemical staining of M30 and γH2AX in in vivo tumor
samples
This study was performed using archived A549 xenograft tumors. These tumors were obtained from mice treated with combination treatment of paclitaxel and carboplatin, followed by single agent suramin or docetaxel or combination treatment of suramin and docetaxel. The animal treatment protocol was described elsewhere (90).
Briefly, subconfluent A549 cells were harvested and implanted subcutaneously into both
flanks of mice (2 × 106 cells per site). After 2 weeks, animals bearing tumors >5 mm in
diameter were used for experiments. A549 tumor-bearing mice received Cremophor
vehicle (untreated control as control of normal tumor growth, n = 6) or 30 mg/kg
paclitaxel intravenously plus 100 mg/kg carboplatin intraperitoneally (n = 20) twice
weekly for two treatments. The P/C pretreated mice were retreated with physiological
saline (pre-treated control as control of pretreatment group, n = 5), 10 mg/kg suramin
(suramin, n = 5), 10 mg/kg docetaxel (docetaxel, n = 5), or docetaxel plus 10 mg/kg
suramin (suramin and docetaxel, n = 5) twice weekly for 3 weeks. Changes in tumor
58 volume and body weight for both treatments were monitored. The name of each group
was stated in parentheses following each treatment.
3.2.7.1 M30 staining
Drug-induced changes in apoptotic cell fraction in tumors were used as pharmacodynamic endpoint for in vivo study. Animal were euthanized 3 days after the
final treatment and tumors were harvested for histological evaluation. Tumors were
excised, weighed and fixed in 10% phosphate-buffered neutral formalin and embedded in
paraffin. Five μm histologic sections were prepared for immunostaining. The apoptotic cells were identified by M30 CytoDEATH antibody (Roche Applied Science,
Indianapolis, IN) and the fraction of apoptotic cells was obtained by counting the number of M30-labeled tumor cells and total tumor cells in 3-5 randomly selected microscopic fields at 200x or 400x magnification, as described elsewhere (91). On average, we counted 1336 ± 787 and 1901 ± 1092 cells per tumor in the pre-treated control (retreated
with physiological saline after pretreatment of paclitaxel and carboplatin) and suramin groups, and 1107 ± 330 and 1499 ± 671 cells per tumor (mean ± SD) in the docetaxel and combination of suramin and docetaxel groups, respectively.
Briefly, tissue sections of paraffin-embedded tumors were deparaffinized, rehydrated, and boiled in 10 mM sodium citrate buffer (pH 6) for 15 min for antigen retrieval. After washing, quenching of endogenous peroxidase activity (1% hydrogen peroxide for 10 min), and blocking, the sections were incubated with M30 antibody at room temperature for 1 hr, followed by incubation with biotinylated secondary antibody
(diluted 1:500) for 30 min and then streptavidinperoxidase complex for 30 min. Positive staining was developed using diaminobenzidine as the chromogen substrate, and sections
59 were counterstained with hematoxylin, dehydrated and mounted using Permount®. For
negative controls, we replaced the primary antibody with the blocking reagent.
3.2.7.2 γH2AX staining
The procedures for immunostaining of γH2AX were basically the same as
immunostaining of M30 described in section 3.2.8. The primary antibody was changed to
anti-γH2AX antibody. The fraction of γH2AX-positive cells was obtained by counting
the number of γH2AX-labeled tumor cells and total tumor cells in 8 randomly selected
microscopic fields at 400x magnification. On average, we counted 1217 ± 558 and 1252
± 514 cells per tumor in the pre-treated control and suramin groups, and 945 ± 351 and
964 ± 306 cells per tumor (mean ± SD) in the docetaxel and combination groups,
respectively.
3.2.8 Statistical Analysis
Statistical significance of the differences between two groups was assessed by the two-tailed Student’s t test. Differences were considered significant when p<0.05.
3.3 RESULTS
3.3.1 Dose response of cisplatin on DSB
To study DNA damage status at various doses of cisplatin, four concentrations were selected, 1, 5, 10, 50 μM, which corresponded to the IC10, IC50, IC80 and IC100 in cultured cells in section 2.3.2, chapter 2. FaDu cells were treated for 1 hr and protein samples were collected at different time points at each dose level and analyzed for
γH2AX amount. Ending time points were decided upon cell confluence condition.
Changes of γH2AX levels over time as a function of cisplatin concentrations are drawn in
60 Figure 3.2. In this chapter, γH2AX expression in all samples was normalized to actin
expression. The results showed little change in γH2AX level at the lower doses of about
or below the IC50 value, followed by a rapid return to the baseline level within 2 or 4
days. At higher cisplatin doses, e.g., IC80 or IC100, the γH2AX levels reached peak
values at day 1 or day 2 after treatment, followed by slower recovery (decline to the baseline level). These data indicate DNA damage intensity and persistence were positively related to cisplatin doses, suggesting γH2AX as a potentially useful PD
endpoint of cisplatin.
3.3.2 Single agent suramin did not enhance DSB
Results of western blotting analysis demonstrated that γH2AX level was not
changed after either low- or high-dose suramin treatment for 72 hr, compared with non-
treated cells (Figure 3.3). Immunohistochemical staining also showed that suramin did
not change γH2AX expression in FaDu monolayer (Figure 3.4).
3.3.3 Addition of suramin to cisplatin altered DSB
Western blot results on the kinetics of γH2AX expression at 1, 2, 4 and 7 days
after cisplatin-alone treatment or combination treatment are given in Figure 3.5. γH2AX
expression induced by single agent cisplatin peaked at day 2, followed by a decline from
day 2 to day 7. Addition of low dose suramin sustained the γH2AX level until day 7, the last day of experiment. This suggests suramin sustained the DSB. In contrast, addition of high dose suramin did not change the γH2AX kinetics compared to single agent cisplatin.
Immunohistochemical staining of γH2AX on FaDu monolayer at day 7 after treatment qualitatively showed more γH2AX expression found in the combination treatment of
61 cisplatin and low dose suramin (Figure 3.6). These results confirmed our hypothesis that
non-cytotoxic suramin enhanced and sustained DNA damage induced by cisplatin.
3.3.4 Cytotoxicity of paclitaxel in FaDu cell lines
The inhibition curve on cell growth of FaDu cells by paclitaxel treatment was
measured by clonogenic assay and shown in Figure 3.7. Cells were treated with paclitaxel
0.1, 0.3, 1, 3, 10, 30, 50, 100, 300, 1000 nM for 24 hr and clonogenicity was calculated
for each concentration, compared with control (containing 0.05% ethanol). Cytotoxicity by paclitaxel was observed from 1 nM and when paclitaxel reached 300 nM, maximum cell killing effects were obtained. The paclitaxel concentration (50 nM) used to study suramin sensitization is close to maximum cytotoxicity.
3.3.5 Alteration of DNA damage by paclitaxel and effect of adding suramin
in FaDu cells
Based on the model shown in Figure 4.1, paclitaxel treatment (50 nM × 24 hr) was combined with pre- and post-treatment of suramin at 20 or 50 μM to study changes in DSB. Results of γH2AX expression in attached cells on day 0, 3, 5 and 7 after paclitaxel treatment with or without suramin in FaDu cells are given in Figure 3.8.
Paclitaxl alone treatment induced γH2AX expression to peak level at day 3 followed by a slow decline at day 7. Single agent suramin did not change γH2AX expression as indicated in section 3.3.2. Two experiments showed qualitatively similar data but quantitatively different data. This may be due to different loading protein amount, exposing time, and developing time. Both experiments showed addition of low dose suramin to paclitaxel slightly raised γH2AX level until day 3 compared to paclitaxel alone treatment, with quickly decreased γH2AX expression to baseline level at day 7.
62 Addition of high dose suramin did not change γH2AX kinetics compared to paclitaxel treatment. Due to high cell detachment rate caused by paclitaxel treatment, we evaluated kinetic expression of γH2AX in detached cells by paclitaxel treatment with or without suramim (Figure 3.9). The results showed there was almost no change on γH2AX level among three treatments. In overall, paclitaxel alone treatment induced γH2AX expression and addition of suramin treatment barely enhanced γH2AX level only around peak time point, indicating γH2AX may not be a good PD endpoint of suramin sensitization to paclitaxel treatment under in vitro condition.
3.3.6 Suramin sensitization and antagonism to paclitaxel in SKOV3 cells
The inhibition curve on cell growth of SKOV3 cells by paclitaxel treatment was measured by clonogenic assay and shown in Figure 3.10. Cells were treated with paclitaxel 0.1, 1, 3, 10, 30, 100, 300, 1000 nM for 96 hr and clonogenicity was calculated for each concentration, compared with control (containing 0.1% ethanol). Cytotoxicity by paclitaxel was observed between 3-10 nM and when paclitaxel reached 300 nM, maximum cell killing effects were obtained. The paclitaxel concentration (20 nM) used to study suramin sensitization is close to maximum cytotoxicity. Single suramin treatment at
20 and 50 μM for 48 hr did not affect SKOV3 cell viability validated by clonogenic assay
(Figure 3.10).
Biphasic effect of suramin on paclitaxel treatment in SKOV3 cell line was observed through clonogenicity measurements (Figure 3.11). Clonogenicity was about
1.6% by 96-hr single paclitaxel treatment at 20 nM. Combination treatment of paclitaxel and suramin was given as suramin was added 24 hr before paclitaxel treatment and continue to treat cells throughout until stopping paclitaxel treatment. Addition of suramin
63 at only 10 μM greatly enhanced chemosensitivity of SKOV3 cells to paclitaxel treatment
by significantly decreasing clonogenicity, which was about half of clonogenicity
achieved by paclitaxel treatment alone. However, pre-treatment of suramin at 20 to 200
μM for the same time length, when combined with same paclitaxel treatment, produced no effect or antagonistic effect as shown by no change or increased clonogenicity. We
chose suramin at 10 and 50 μM to be combined with paclitaxel treatment to study alteration of DNA damage by paclitaxel and effect of adding suramin in SKOV3 cells.
3.3.7 Alteration of DNA damage by paclitaxel and effect of adding suramin
in SKOV3 cells
Paclitaxel treatment (20 nM × 96 hr) was combined with pre-treatment of suramin at 10 or 50 μM to study changes in DSB. Suramin was added 24 hr before paclitaxel treatment and maintained in medium until specific time points of measurement after paclitaxel treatment. Both attached cells and detached cells were collected at each time point. Results of γH2AX expression on day 1, 2, 4 and 7 after paclitaxel treatment with or without suramin in SKOV3 cells are given in Figure 3.12. γH2AX expression was quickly induced in first two days by paclitaxl alone treatment and reached a plateau level at day 4 through day 7. Addition of low dose suramin to paclitaxel raised γH2AX level compared to paclitaxel alone treatment, with the same kinetics pattern. Addition of high dose suramin had higher initial level of γH2AX than single agent paclitaxel and reached plateau level at day 2. The results showed there were similar γH2AX kinetics changes
among three treatments. This observation was consistent with results in section 3.3.5 and
confirmed that γH2AX may not be a good PD endpoint of suramin sensitization to
paclitaxel treatment under in vitro condition.
64 3.3.8 Apoptosis by docetaxel and enhancement by addition of suramin
under in vivo setting
To establish an in vivo model to investigate the role of γH2AX as PD endpoint to suramin sensitization to non-DNA-damaging treatment, we used docetaxel, a paclitaxel analog, to combine with suramin and treated A549 xenograft tumors after pretreatment of
paclitaxel/carboplatin. As shown in Figure 3.13 and Figures 3.14, single-agent suramin
had no effect on apoptosis in tumors compared to pre-treated controls, as indicated by a
similar percentage of apoptotic cells. Single-agent docetaxel significantly increased the
apoptotic index (p<0.05 compared to paclitaxel/carboplatin pre-treated control and single
agent suramin). The addition of low dose suramin significantly increased the apoptotic
index over docetaxel alone treatment (p<0.05).
3.3.9 Evaluation of DNA damage by docetaxel and alteration by addition of
suramin under in vivo setting
Immunostaining of γH2AX gave similar values among four treatments regarding
percentage of γH2AX-positive cells (Figure 3.15, and 3.16). Again, suramin alone did not
induce γH2AX, reflected by a similar percentage to paclitaxel/carboplatin pre-treated
controls. Single agent docetaxel had percentage of positive cells close to pre-treated
control as well. This index was not affected by the combination of suramin and docetaxel,
compared with pre-treated control, and was increased not significantly, compared with
single agent docetaxel. This suggests γH2AX may not serve as PD endpoint of suramin
sensitization to docetaxel treatment under in vivo condition.
65 3.3.10 Summary
All collective data in this chapter suggested γH2AX might be used as PD point of
suramin sensitization to DNA-damaging treatment such as cisplatin, but may not be a
good PD endpoint to non-DNA-damaging treatment, e.g. paclitaxel.
3.4 DISCUSSION
Double strand breaks can be induced by a variety of chemotherapeutics and ionizing radiation. These agents induce cytotoxicity through involving different biochemical pathways and stimulate γH2AX expression at specific timing. Formation of
γ-H2AX foci were detectable within minutes of radiation treatment and reached plateau quickly (13; 25). Topoisomerase II inhibitors such as doxorubicin and etoposide, and
DNA strand-break inducer bleomycin, maximized γH2AX expression within hours after treatment (14). In our study, cisplatin-induced γH2AX appeared relatively slowly, particularly in 1 to 2 days. This observation may be due to the fact that DSB production arises through repair of DNA-cisplatin adducts and crosslinks, rather than from direct
DNA damage provoked by other DNA damaging treatments such as ionizing radiation.
The γH2AX expression induced by cisplatin is dose-dependant as observed by many studies in ionizing radiation (13) and DNA-damaging agents (14). The numbers and intensity of γ-H2AX foci were quantitatively correlated well with increased dosing in ionizing radiation (92; 93). For topoisomerase I and II inhibitors, γH2AX levels increased
in direct proportion to drug dose (14). Relationship between cisplatin dose and peak
amount of γH2AX expression detected in our western blotting analysis revealed a strong
association. There might be a threshold dose of cisplatin for γH2AX to be induced.
66 Further experiments are needed to confirm this correlation with more sensitive method, such as flow cytometry.
Few studies focused on persistence of γH2AX expression after DNA-damaging treatment in a range of doses. Our data showed clearance of γH2AX was at much slower rate by cisplatin at toxic dose. This could be reasoned that high dose cisplatin produced more DNA damage and needed more time to repair, as γH2AX is a possible marker of unrepaired DNA damage (13). Therefore, a number of studies explored linkage of
γH2AX removal to tumor sensitivity in radiation treatment or chemotherapy (14; 94; 95).
Our study is among the first to report that repair kinetics of γH2AX correlated with the dose response of cisplatin.
The influence on tumor sensitivity of low- and high-dose suramin to cisplatin of various doses was demonstrated in a pilot study. Considering the severity of DNA damage by a range of cisplatin doses, suramin at 20 or 50 μM was combined with cisplatin at 5, 10, 20 or 30 μM. Relative ratio of combination treatment in clonogenicity was calculated over single cisplatin treatment at each dose. It showed that while antagonism by high dose suramin was not changing much at higher cisplatin doses, sensitization by low dose suramin was more significant when cisplatin dose was increased. Considering enhanced DNA damage by high cisplatin dose, suramin sensitization might be achieved through biochemical mechanism related to DNA damage change.
Based on all cellular PD observations regarding suramin sensitization and antagonism to cisplatin and relationship of γH2AX as marker of tumor sensitivity in cisplatin treatment, we first observed γH2AX disappearance was prolonged by addition
67 of low dose suramin with enhanced tumor sensitivity. Because cisplatin at 30 μM is very
toxic dose and produces intensive DNA damage, suramin sensitization to cisplatin at this
dose is found due to DNA damage status change, or more specifically, persistence change
of DNA damage, rather than intensity change. This interesting finding would lead to
more studies on mechanism of biphasic effect of suramin and explore usage of γH2AX as
PD endpoint in combination treatment with suramin.
Yu et al. has found that γH2AX can be endogenously expressed in untreated tumor cells (96). As a consequence of genomic instability, it is observed that the expression level of endogenous γH2AX was higher in p53 mutated cells than p53 wild- type cells (96). Since FaDu cells used in our model are p53 mutated, the observable base level of γH2AX expression is found in this cell type by our western blotting bands.
In multiple experimental models, non-toxic suramin sensitized to a number of chemotherapeutics. Our finding that γH2AX might be PD endpoint of suramin sensitization to cisplatin treatment, raised an intriguing question: does it apply to non-
DNA-damaging treatment? We chose paclitaxel combined with suramin, which showed similar biphasic effect as well in FaDu and SKOV3 cell lines using clonogenic assay.
Change of γH2AX level with time was different in two cell lines by single agent paclitaxel. This cell type dependency might be related to treatment schedule, dose, and genetic difference in cells. Both models showed that kinetics change of γH2AX were similar by addition of suramin compared with paclitaxel alone treatment. In FaDu cells, addition of low dose suramin showed the biggest difference at day 3 from single paclitaxel treatment. We then evaluated γH2AX change at day 3 after the final dose under in vivo conditions, showing very low γH2AX induction by paclitaxel alone treatment or
68 combination treatment compared with base level in untreated cells. Though combination treatment with low dose suramin had more γH2AX-positive cells than single agent paclitaxel, which was consistent with in vitro observation, the difference was not significant statistically. These results suggested that γH2AX may not be feasible PD endpoint in suramin sensitization to taxanes.
69 8 )
6 *
* 4
* 2 * * Clonogenicity (% of control of (% Clonogenicity 0 0 5 10 20 30 40 50 100 Suramin concentration (uM)
Figure 3.1 Suramin sensitization and antagonism to paclitaxel treatment (50 nM × 24 hrs) in FaDu cells. * indicates P<0.05 compared to paclitaxel alone treatment (by student’s t test). Data are mean + SD (n=3 replicates in 1 experiment). The bar represents one SD. Data was provided by Dr. Mingjie Liu.
70 A.
Experiment 1
4000 Cis_1uM n Cis_5uM 3000 Cis_10uM Cis_50uM
2000 (% of control) 1000 Relative rH2AX/Acti ratio of
0 04812 Days after cisplatin treatment
Continued
Figure 3.2 Kinetic change of γH2AX expression (normalized to actin expression) after one-hour cisplatin treatment at different doses in FaDu monolayer. A. Data from two experiments were shown. Ending time points were decided upon cell confluence condition. Due to different conditions such as loading protein amount, exposing time, and developing time, two experiments had quantitatively different but qualitatively similar data. B. Western blotting images of γH2AX and actin expression after cisplatin treamtment at different doses. 0 day means samples collected right after cisplatin treatment.
71 Figure 3.2 Continued
Experiment 2
600 Cis_1uM Cis_5uM 500 Cis_10uM 400 Cis_50uM
300
(% of control) of (% 200
100 Relative ratio of rH2AX/Actin 0 04812 Days after cisplatin treatment
B.
Cis1μM Cis5μM __ _ Cont 0d 1d 2d 0d 1d 2d 4d γH2AX Actin Cis10μM Cis50μM _ __ Cont 0d 1d 2d 4d 6d 0d 1d 2d 4d 7d 12d γH2AX Actin
72 A.
150
120
90
60 (% of control) (% of 30
Relative ratio of rH2AX/Actin rH2AX/Actin of ratio Relative 0 02050 Suramin concentration (uM)
B.
Cont S20 S50 γH2AX Actin
Figure 3.3 A. Ratio of γH2AX/actin unchanged by suramin-alone treatment for 72 hr.
There was no significant difference among treatments. Data are mean + SD (n=3
experiments with a single observation per experiment). The bar represents one SD. B.
Western blotting images on γH2AX and actin expression after suramin-alone treatment.
73 Control Suramin 20 μM
Suramin 50 μM
Figure 3.4 Immunohistochemical stainning of γH2AX by suramin-alone treatment on day 7. These results qualitatively showed cellular location of γH2AX (n=3 replicates in 1 experiment).
74 A.
Experiment 1 Experiment 2
800 6000 Cis30 Cis30 Cis30 + S20 5000 Cis30 + S20 600 Cis30 + S50 Cis30 + S50 4000
3000 400
2000 control) (% of (% of control) of (% 200 1000 Relative ratio of rH2AX/actin Relative ratio of rH2AX/actin 0 0 02468 02468 Days after cisplatin treatment Days after cisplatin treatment
B.
Cis30 Cis30+S20 Cis30+S50 . Cont 1d 2d 4d 7d 1d 2d 4d 7d 1d 2d 4d 7d γH2AX Actin
Figure 3.5 A. Kinetic change of γH2AX expression (normalized to actin expression) after cisplatin treatment with or without suramin in FaDu monolayer. Data from two experiments were shown. Due to different conditions such as loading protein amount, exposing time, and developing time, two experiments had quantitatively different but qualitatively similar data. B. Western blotting images on γH2AX and actin expression after three treatments.
75 Cisplatin 30 μM Cisplatin 30 μM + Suramin 20 μM
Cisplatin 30 μM + Suramin 50 μM
Figure 3.6 Immunohistochemical stainning of γH2AX by cisplatin treatment with or without suramin on day 7. These results qualitatively showed cellular location of γH2AX
(n=3 replicates in 1 experiment).
76 120
100
80
60
40
20
Clonogenicity (% of control) (% of Clonogenicity 0 0.1 1 10 100 1000 Paclitaxel concentration (nM)
Figure 3.7 Cytotoxicity of 24-hr paclitaxel treatment in FaDu monolayer culture. Drug effect was measured using the clonogenic assay. Data are mean + SD (n=3 replicates in 1 experiment). Some SDs are smaller than the symbols. The clonogenicity in the untreated
control was 95.6+5.7%.
77 A.
Experiment 1 Experiment 2
600 300 TX50 TX50 500 250 TX50 + S20 TX50 + S20 TX50 + S50 TX50 + S50 400 200
300 150
200 (% of control) of (% 100 (% of control) (% of
100 50 Relative ratio of rH2AX/actin Relative ratio of rH2AX/actin 0 0 02468 02468 Days after paclitaxel treatment Days after paclitaxel treatment
B.
TX50nM TX50+S20______Cont 0d 1d 2d 3d 5d 7d 0d 1d 2d 3d 5d 7d γH2AX
Actin TX50 TX50+S50______Cont 0d 0d 1d 2d 3d 5d 7d γH2AX Actin
Figure 3.8 A. Kinetic change of γH2AX expression (normalized to actin expression) after paclitaxel treatment with or without suramin in FaDu monolayer (attached cells).
Data from two experiments were shown. Due to different conditions such as loading protein amount, exposing time, and developing time, two experiments had quantitatively different but qualitatively similar data. B. Western blotting images on γH2AX and actin expression after three treatments.
78 A.
3000 TX 2500 TX+S20 TX+S50 2000
1500
(% of control) of (% 1000
500 Relative ratio of rH2AX/actin
0 02468 Days after paclitaxel treatment
B.
TX50nM TX50+S20 Cont 1d 2d 3d 5d 7d 1d 2d 3d γH2AX Actin TX50+S20 TX50+S50______Cont 5d 7d 1d 2d 3d 5d 7d γH2AX Actin
Figure 3.9 A. Kinetic change of γH2AX expression (normalized to actin expression) after paclitaxel treatment with or without suramin in detached FaDu cells. B. Western
blotting images on γH2AX and actin expression after three treatments.
79
Figure 3.10 A. Cytotoxicity of 96-hr paclitaxel treatment in SKOV3 monolayer culture.
Drug effect was measured using the clonogenic assay. Data are mean + SD (n=3
replicates in 1 experiment). Some SDs are smaller than the symbols. The clonogenicity in
the untreated control was 60.9+8.3%. B. Cell viability by suramin alone treatment (20 or
50 μM × 48 hr) on SKOV3 cell line. There was no significant difference compared to
untreated control. Data are mean + SD (n=3 replicates in 1 experiment). The bar
represents one SD. The clonogenicity in the untreated control was 101.3+10.0%.
80 Figure 3.10
A.
120
100
80
60
40
20 Clonogenicity (% of control) 0 0.1 1 10 100 1000 Paclitaxel concentration (nM)
B.
150
100
50
Clonogenicity (% of control) of (% Clonogenicity 0 02050 Suramin concentration (uM)
81
Paclitaxel Paclitaxel (20 nM ×96 hr) combined with suramin only 4
3 *
2
1 * Clonogenicity (% of control) (% of Clonogenicity 0 0 1 5 10203050100200 Suramin concentration (uM)
Figure 3.11 Suramin sensitization and antagonism to paclitaxel treatment (20 nM × 96 hr) in SKOV3 cells. * indicates P<0.05 compared to paclitaxel alone treatment (by student’s t test). Data are mean + SD (n=2 experiments with 3 replicates per experiment).
The bar represents one SD. The clonogenicities in the untreated control were 91.7+9.1%
and 59.4+4.3%, respectively.
82 A.
5000
4000
3000
2000 TX20 (% of control) (% of TX20 + S10 1000 TX20 + S50 Relative ratio of rH2AX/Actin rH2AX/Actin of ratio Relative 0 02468 Days after paclitaxel treatment
B.
TX20nM TX20+S10 TX20+S50 . Cont 1d 2d 4d 7d 1d 2d 4d 7d 1d 2d 4d 7d γH2AX Actin
Figure 3.12 A. Kinetic change of γH2AX expression (normalized to actin expression) after paclitaxel treatment (20 nM × 96 hr) with or without suramin in SKOV3 cells (both attached and detached cells). Data from one experiment were shown. B. Western blotting
images on γH2AX and actin expression after three treatments.
83 6
5
4 ** 3
2 *
% M30-positive cells 1
0 Pre-treated Suramin Docetaxel Docetaxel + control Suramin
Figure 3.13 In vivo apoptosis of docetaxel/suramin in A549 xenograft tumors after pretreatment of paclitaxel/carboplatin.
Mice received treatments as described in section 3.2.7. Three days following administration of the final dose, tumors were harvested and fixed in formalin.
Histological sections were stained for M30, a marker of apoptotic cells. The apoptotic index was calculated as (number of M30-stained cells) divided by (total cell number) in
3-5 “hotspot” microscopic fields at 200× or 400× magnifications in each tumor. Eight to eleven tumors per group. Mean + SD. * indicates P<0.05 compared to pre-treated control and single-agent suramin. ** indicates P<0.05 compared to single agent docetaxel.
84 Pre-treated control Suramin
Docetaxel Suramin + Docetaxel
Figure 3.14 Micrographs showing M30 immunohistochemical staining in A549 xenograft tumors after each treatment (400× magnifications, n = 8-11).
85 1.2
0.9
0.6
0.3 % rH2AX-positive cells
0.0 Pre-treated Suramin Docetaxel Docetaxel + control Suramin
Figure 3.15 In vivo γH2AX expression of docetaxel/suramin in A549 xenograft tumors after pretreatment of paclitaxel/carboplatin.
Mice received treatments as described in section 3.2.7. Three days following administration of the final dose, tumors were harvested and fixed in formalin.
Histological sections were stained for γH2AX, a marker of DNA damage. The percentage of γH2AX-positive cells was calculated as (number of γH2AX-stained cells) divided by
(total cell number) in 8 “hotspot” microscopic fields at 400× magnifications in each tumor. There was no significant difference between single suramin or docetaxel treatment and pre-treated control, or between combination treatment and single docetaxel treatment.
Two to five tumors per group. Mean + SD.
86 Pre-treated control Suramin
Docetaxel Suramin + Docetaxel
Figure 3.16 Micrographs showing γH2AX immunohistochemical staining in A549 xenograft tumors after each treatment (400× magnifications, n = 2-5).
87
CHAPTER 4
SEROLOGIC RESPONSE TO HUMAN PAPILLOMAVIRUS TYPES 6, 11, 16
AND 18 IN CHINESE WOMEN
4.1 INTRODUCTION
Numerous biological and epidemiological studies support the etiological role of human papillomavirus (HPV) infection in cervical neoplasia (97-99). Studies using
different methodological approaches have detected oncogenic HPV DNA in nearly all
cases of cervical cancer and 80-90% of high-grade dysplasia by the most sensitive and
specific DNA detection techniques (100-102). HPV 16 is the most common oncogenic
HPV type and presented tactable in about half of cervical cancer biopsy tissue (103).
HPV 18 has been followed to be the second most common type in invasive cervical
cancer in China (104) and the globe regions.
Condylomata acuminata, also known as genital warts, is the most common benign
tumor in the anogenital region (105). Together, HPV types 6 and 11 are the main causal
agents and have been detected in up to 83% of genital warts (106) in Mainland China.
HPV 6 has been detected approximately three times more often than HPV 11 in these
lesions (107).
88 Recent advances have been made in detecting HPV exposure through antibody responses. HPV DNA is transient in exfoliated cell or tissue and thus it cannot provide a reliable indicator of past exposure. HPV DNA detection is also limited by sampling difficulties because many unmarried women are unwilling to undergo gynecologic examinations for the collection of exfoliated cells due to culture difference in China.
HPV antibody responses are a useful marker of cumulative exposure to HPV.
Researchers have identified virus-like particles (VLPs) that make it possible to detect
HPV antibodies potentially indicative of previous exposure to HPV infection. Several studies have detected antibodies to neutralizing epitopes on virus-like particles for HPV types 6, 11, 16, and 18 by multiplex Luminex assay (108; 109).
Few studies have estimated HPV seroprevalence using representative, population based samples. Additionally, limited data are available concerning the usefulness of HPV antibodies as a marker in cervical cancer screening. There is also a lack of population- based data describing cumulative HPV exposure along time in Chinese women.
Therefore, in this investigation, we measured seroprevalence of four common HPV types
(6, 11, 16, and 18) from five regions of Mainland China. We also studied the presence of
HPV antibodies in patients with cervical neoplasia diagnosed by pathologists or detected by cytology or visual inspection after acetic acid (VIA) method.
4.2 MATERIALS AND METHODS
4.2.1 Study population
This cross-sectional study was performed within Mainland China, and included study subjects from 5 areas of China, Shanxi (north, suburban), Beijing (north, urban),
89 Xinjiang (west, suburban), Henan (north, suburban) and Shanghai (south, urban).
Subjects aged 15 to 54 were included in this study. We notified women with booklets,
notices placed on community bulletin boards, and television announcements. Women
were also invited to participate in this study by village doctors who visited each
household in the community with eligible women. Women who agreed to participate
signed their names on the consent form or put their thumbprint. The consent form was
approved by the Ethics Committees of Cleveland Clinic and Cancer Foundation of China.
Consenting women were enrolled in an age-stratified quota sample, with a maximum of 125 women in each of eight 5-year age strata. When the quota ceiling was
exceeded, women were offered the free standard pap. However, we encountered
difficulties in recruiting women at age of 15 to 19, because most of women in this age
group are not married yet and were not willing to undergo gynecological examination.
Exclusion criteria consisted of women reporting a hysterectomy, history of radiotherapy
or current pregnancy in the study interview.
4.2.2 Sample collections
Nine ml of blood was taken from all consenting females. Blood samples were
centrifuged at 1,500×g for 10 min and aliquoted into plasma, buffy coats and RBC.
Aliquots of blood were stored at -80 . All of the included women who were sexually
active consented to provide physician-collected exfoliated cervical cells used for HPV
DNA detection by HC2 (hybrid capture 2, QIAGEN Corp, Gaithersburg, MD). All HC2
positive samples and 5% of HC2 negative samples randomly drawn were genotyped by
Linear Array PCR assay (Roche). Liquid Based Cytology (SurePath, Becton Dickinson,
Franklin Lakes, NJ) was also performed using cervical cells. Exfoliated genital cells were
90 stored at room temperature. A threshold value of at least 1.0 pg/ ml HPV DNA in HC2
test was used as the cut-off for positives.
VIA test was performed after cervical cell collection. Women with positive VIA
received direct biopsy under colposcopy at the first visit. Those with negative VIA but abnormal cytology or high-risk HC2 positive results were recalled for direct or random biopsies during colposcopic examination. Endocervical curettage was done if indicated.
Final diagnoses on biopsy tissue were made by pathologists at CICAMS. All of biopsy
tissues of cervical intraepithelial neoplasia (CIN) 2 or above were genotyped by Linear
Array (Roche) for specific HPV types.
4.2.3 HPV antibody detection by multiplex Luminex assay
Yeast-derived VLPs are coupled to a set of four distinct fluorescent Luminex
microspheres by using conjugation chemistry as previously described [13]. Antibody
titers were determined in a competitive format in which known, type-specific
phycoerythrin (PE)-labeled, neutralizing monoclonal antibodies (mAbs) compete with the
subject’s serum antibodies for binding to type-specific, conformationally sensitive,
neutralizing epitopes on the VLPs. The fluorescent signals from the bound HPV-specific
detection mAbs are inversely proportional to the subject’s neutralizing antibody titers.
Results for the assay are reported as concentration of antibody in arbirtrary milli-Merck
Units per milliliter (mMU/mL).
The HPV 6, 11, 16 and 18 complex Luminex immunoassay (cLIA) is performed in a 96-well microtiter plate. A 12-point standard reference serum from hyper-immunized
African green monkeys, 4 controls and 16 samples are added to the plate in duplicate.
Samples are tested at a 1:4 dilution. To each well is added the detection antibodies
91 followed by the VLP-microspheres for types 6, 11, 16 and 18. The plates are sealed with
foil covers and incubated overnight for 15 to 25 hours. Following incubation, the plates
are washed 3 times and then analyzed on a BioPlex (Luminex) instrument. The high,
medium, low and negative controls used for this assay were collected from humans that
were either HPV sero-negative, had low antibody titers from natural infection or had
medium to high Ab titers to HPV L1 VLPs following vaccination. The details of cLIA
was described somewhere else (108; 109).
4.2.4 Statistical analysis
Database was established and incorporated using Microsoft Visual FoxPro
(version 8.0). Basic statistical analysis was done with R (version 2.8.0). Odds ratios (OR)
for HPV seropositivity and corresponding 95% confidence intervals (CI) were calculated
by nonconditional logistic regression model, with adjustment for age and HPV DNA
status (STATA version 9.0).
4.3 RESULTS
4.3.1 Statistics of study subjects
A total of 4,731 females with valid HPV serology results were included in this
study and divided into two groups according to their sexual history. In the first group,
there were 4,211 subjects with sexual history aged from 17 to 54 (median 37). Each site has about equal number of subjects enrolled. Women at Xinjiang site are mostly from
Uyghur ethnic group, whereas other sites represent majority ethnic group, Han group.
One third of women have an elementary or lower education level (32.2%). Most of women are married (94.8%), never smokers (96.9%) and ever use contraceptive method
92 (89.5%). Most women (79.9%) had started their sexual life before 25 years of age, and
had had a single lifetime sexual partner (77.3%). The second group was subjects without sexual history and totally there were 520 subjects aged from 14 to 36 (median 18).
4.3.2 HPV seropositivity in sexually active women and virgins
Overall and type-specific HPV seroprevalence is shown in Figure 4.1, stratified
by sexual activity status. The decreasing order of seropositive HPV types was HPV 6, 16,
11 and 18, which was the same case for both sexually active and naive subjects. HPV
seroprevalence was significantly higher among sexually active (15.8% for any HPV type)
than virgins (2.5%), and this excess was significant for each tested HPV type. The strong
association between HPV seropositivity and sexual activity was also seen in women with
increasing number of sex partners or with their husband’s having extramarital sexual
relationship, shown as below.
4.3.3 Age-specific sero-prevalence in Chinese women
To show the changing trend of HPV seroprevalence with age, we calculated
sexually active women’s age-specific antibody prevalence rate. Overall and type-specific
HPV 6, 11, 16 and 18 seroprevalence are shown in Figure 4.2. Because the number of subjects in age group of 15-19 was too low (n=27), it was combined into age group of 20-
24 to be comparable to other age groups. Age-specific seroprevalence of HPV 11 and
HPV 18 are constantly low in different age groups with a bit higher prevalence of HPV
11 (2.2 – 3.9%) than that of HPV 18 (1.3 – 2.9%). HPV 16 showed low serological prevalence in younger women under 30 years old and increased serological responses in older women after 30 to the level of around 7%. The highest prevalence was that of anti-
HPV 6 antibodies. It presented a plateau (7.5 – 7.8%) until age of 40, began to climb then
93 and peaked (9.8%) at late forties, with decreased prevalence in women after 50 years old.
Therefore, the overall seroprevalence in Chinese women gave a gradually increasing
trend along with the age, with a peak (19%) showed at age of 45 to 49 and a quick
decline followed.
4.3.4 Association between sociodemographic and sexual characteristics and
seropositivity
The association between several selected sociodemographic characteristics and
sexual characteristics and seropositivity for HPV 6/11/16/18, HPV 6 and HPV 16 antigens was shown in Table 4.1. Female aged under 35 years old had a lower risk of
being seropositive for HPV 6/11/16/18 antigen compared to women aged 35 to 44 (OR =
1.40; 95% CI: 1.03-1.91) or women aged 45 to 54 (OR = 1.40; 95% CI: 1.02-1.90), when
controlling for HPV DNA positivity. Such difference of seropositivity risk at various
ages is possibly due to higher risk of being seropositive for HPV 16 antigen in older
women, instead of HPV 6 antigen. Uyghur minority women were more likely to be
seropositive for HPV 6 antigen (OR = 1.40; 95% CI: 1.08-1.83), compared to Han
majority women. Women from rural regions had a significantly lower seropositivity for
HPV 6/11/16/18 (OR = 0.72; 95% CI: 0.60-0.87) antigen compared with women from
metropolitan area. With the increasing number of sex partners, women were more likely
to be seropositive for all kinds of HPV types. However, the increasing trend was slightly reversed in women having three sex partners to be seropositive for HPV 16 antigen (OR
= 1.40; 95% CI: 0.80-2.46). Meanwhile, wives were at higher risk of exposure to HPV antigen of all types when their husband had extramarital sexual relationship (OR = 1.80;
95% CI: 1.46-2.22). There is no difference of seropositivity for HPV 6/11/16/18 antigen
94 in females regarding their education level, marital status, smoking or drinking habits, parity number, or among women whose age at first menstruation or sex intercourse is different, or with usage of contraceptive measures or not. Women who had induced abortion ever were more likely to be positive for antigen against any of four types, though the trend is not statistically significant. There are significantly high anti-HPV 16 in women having history of induced abortion (OR = 1.74; 95% CI: 1.03-2.93).
4.3.5 Seropositivity in women having normal cervical or with cervical
lesions
Table 4.2 reported HPV 6/11/16/18, HPV 16 and HPV 18 serological antibody expression in women having normal cervix or cervical intraepithelial lesions. In women with high-grade cervical lesions, antibody level against HPV 6/11/16/18 antigen were much higher than that in women with normal cervix diagnosed by cytology (OR = 3.69;
95% CI: 2.12-6.43), pathology (OR = 3.74; 95% CI: 2.21-6.33) and VIA method (OR =
1.74; 95% CI: 1.06-2.85). Women diagnosed with high-grade cervical lesion by cytology and pathology methods were also more likely to have higher anti-HPV 16 in serum than women with normal cervix, but this difference was not significant regarding lesion diagnosis by VIA method. We did not find the same pattern of anti-HPV 18.
4.3.6 Concordance between HPV DNA positivity and seropositivity
Comparison of seropositivity and HPV DNA positivity detected by HC2 test was given in Table 4.3. Only about one fifth (21.18%) of women with HPV DNA of high-risk types were detected to have antibody against HPV 16 or 18. This percentage (16.95%) was even lower in women with anti-HPV 6 or 11 detected when HPV DNA of low-risk types was positive. Due to inability to genotype HPV DNA by HC2, these relatively low
95 percentages were in the possible range. For women with no HPV DNA of high- or low- risk detected, less than one tenth (5.82% (210/3611), 8.61% (311/3611)) of women had antibody against HPV 16/18 or HPV 6/11, suggesting antibody persistence after DNA clearance in these women. Nevertheless, women with positive DNA were more likely to be seropositive than women with negative DNA results.
To gain better understanding about relationship between HPV DNA and antibody positivity and its implication in practice, Table 4.4 presented the HPV seropositivity stratified by HPV type, HPV DNA positivity detected by Linear Array and pathology status. In women with either low- or high-grade cervical intraepithelial neoplasia (CIN), around half of women (47.37%, 57.45%, respectively) with presence of HPV 16 DNA were seropositive for HPV 16 antigen. One noteworthy fact is that anti-HPV 16 was detected in about one forth (23.81%) of women having CIN2 or above while negative for
HPV 16 DNA. It indicates that persistence of HPV 16 infection or antibody expression is related to cervical neoplasia development. In contrast, anti-HPV 18 was less likely
(14.29%, 16.67%) to be present in women with any grade of cervical neoplasia when their cervix were infected with HPV 18. As non-oncogenic types, HPV 6 or 11 DNA were almost absent in cancerous tissue. However, regardless of pathology status, the total number of women being seropositive for HPV 6 or 11 antigens is much more than that of women with HPV 6 or 11 DNA present in cervix (22 vs. 3 in CIN 1, 15 vs. 2 in CIN 2 or above), suggesting persistent existence of anti-HPV 6 or 11 in serum after virus infection and clearance.
96 4.4 DISCUSSION
As HPV is sexually transmitted virus, the relationship between HPV
seroepidemiology and sexual activities is in our interest. A number of studies have found
that seropositivity of HPV 16 (110-113), 18, and 11 (113) is associated with number of
sexual partners in life. Linear trend was reported in some studies with increasing number
of sexual partners (110; 112; 113). In our study, we observed seropositivity of HPV 6, 11,
16 and 18 in sexually active women is consistently statistically higher than in virgins. In
sexually active women, serological response is not evidently in linear association with
increasing number of sexual partners for any HPV type (data of HPV 11 and 18 not
shown). Seroprevalence of HPV 6 or 16 was the lowest among those with one partner,
and increased abruptly with two partners and slowly reached the plateau until more than
four partners. This plateau phenomenon was also observed in study by Olsen et al. (114),
in which with more than three partners, the plateau was reached. This suggests sexual
activity might be related to not only number of sexual partners, but also other factors such
as sexual life of their partners found in other study (115). Indeed in our study we found
that husband’s sexual activity is a significant covariate to HPV seropositivity of wives for
any HPV type. Though husbands’ extramarital relationship is provided by wives in
questionnaire and might be inaccurate, the association between seropositivity and
husbands’ affairs is still strong and indicates that “male factor” is an important source of
cumulative exposure of HPV to women. One interesting observation is that for women with three sexual partners, antibody response of HPV 11, 16, and 18 were all at lower percentage than those with two or four partners. This may be because of the smaller size number of this group. Seroprevalence among virgins was low though detectable,
97 compared to other studies ((116; 117), 2-3% for HPV 16 in virgins). Andersson-Ellström
et al. reported no seropositivity among virgins in their study (118). Except possible nonsexual transmission or infection at other sites that could induce seroconversion (119),
sexual activity was self-reported by virgins in our survey. Therefore, caution should be
taken when interpreting results in virgins.
As the first report of population-based prevalence of HPV antibody in Chinese
women, our investigation showed that there is relatively low seroprevalence of HPV 16
and 18 among Chinese females, compared to studies in other areas worldwide (52). Anti-
HPV 16 and 18 prevalence in our study is most comparable to a population-based study
in Taiwan (6.3% vs. 7.6%, 2.1% vs. 3.9%) (120), but much lower than in Mongolia
(6.3% vs. 23%, 2.1% vs. 19.6%) (115). Seroprevalence of HPV 16 is at similar level in
China and in South Korea (6.3% vs. 6.3%), whereas anti-HPV 18 is less prevalent than in
South Korea (2.1% vs. 9.0%) (121). Since HPV 16 is dominant type in cervix DNA samples in China (122; 123), seroprevalence of HPV 16 is found consistently more prevalent than HPV 18 in all areas in China, considering a common sexual route of transmission by all HPV types. HPV 6 turns out to be the most prevalent sero-type found in our study, which is also observed in a similar study in United States (124). Though cross-reactivity between HPV 6 and HPV 11 cannot be ruled out due to many conservative sequences shared by these two types (109), seroprevalence of HPV 6 remains much higher than other three types. Because HPV 6 is not among the most common types in HPV infection in China (122), we speculated a high possibility of cumulative exposure to HPV 6 in China, though the specific reason is unknown. Other possibilities include distinct immunogenicity stimulated by HPV 6 over other HPV types
98 (125). Multiple positivity is not high between HPV 6 and 11 or 16 and 18 (data not
shown). It might be because serum antibody response is a marker of lifetime cumulative
exposure and cross protection between types might exist (126). Given big population
base and relatively conservative sexual behavior in China in the past years, prevalence of serological response to HPV is as expected to be low in a diverse range of seroprevalence
globally.
The age trend of HPV seropositivity in our study gave a peak for anti-HPV of any
type at late 40s and a relatively flat curve for individual HPV types at different age group.
On the other hand, high-risk HPV DNA prevalence peaked at both early twenties and
early forties in Chinese women (123) and low-risk HPV DNA has very low prevalence in
China. Although we do not have male seroprevalence data in our study, data in Taiwan
showed a second peak at older age in men and a second peak of HPV DNA observed in
women in Mainland might suggest the increasing trend of age-specific prevalence.
Joakim reviewed that viral load and persistence are two most important variables to HPV
seropositivity (127). The peak of seroprevalence appeared after the second peak of DNA
prevalence, but not after the first peak. Combined with the lag time longer than time
needed for seroconversion to occur, it suggests repeated or recurrent infection (128), or
reactivation of latent infection composed of continued exposure to women at older ages
and such persistent exposure is necessary to simulate serological response. Cohort effect
and low seroconversion rate should also not be ignored. The sudden decline in
seroprevalence in women after late forties, observed in other studies as well (116; 120;
129; 130), might be because of waning immune response in older people. From the
perspective of prevalence, women before 30 had a relatively low seroprevalence,
99 comparable to a study of young women in South Korea (131), Thailand (129) or Costa
Rica (130). Age-specific seroprevalence of HPV 16/18 showed similar trend to that in
Mongolia (115) and more steady change than in Taiwan (120). The age range in which
HPV 16 seroprevalence peaked in China was much later than in US (45-49 vs. 25-29), but for other three types falls in about the same range. In overall, age-specific prevalence in China showed similar trend to other areas, with steadily increasing tendency with age.
When investigating regional difference on HPV seroprevalence, we found that big difference exists not only between rural and metropolitan areas, but also within metropolitan areas. There was significantly higher seropositivity of any HPV type, particularly HPV 6 or 16, in Shanghai than in Beijing (OR = 1.57; 95% CI: 1.21-2.04).
Furthermore, fewer women with one lifetime sexual partners and more women with two partners were recruited in Shanghai than in Beijing. But number of women with three or more sexual partners is comparable between two biggest metropolitan cities in our study.
On the other hand, there is no difference on seropositivity of these four types between two rural areas, Xinmi county of Henan Province and Xiangyuan county of Shanxi
Province. Distribution of sexual activity among Uyghur women shifted to have more number of sexual partners than Han women from four above areas. Therefore, antibody level against HPV 6 antigen is significantly high in Uyghur women, but not for antibody level against HPV 16 antigen. Generally, there is regional and ethnicity difference among different geographical sites, in consideration of sexual behavioral difference.
Education level, smoking, and drinking has no impact on seropositivity in our study, consistent with other studies (115; 130) though Shin et al. found association exists in their study (121). Married women have lower seropositivity than single or divorced
100 women or widows, possibly due to more stable sexual relationship among married women. Among menstrual and reproductive factors, women with pregnancy history are more likely to be seropositive for HPV 16 antigen than women never pregnant before, but not for other types. The detection that pregnancy history was related to higher anti-HPV
16 is different from what have been found in other studies (115; 121). Induced abortion might be cofactor of number of pregnancies or number of lifetime sexual partners and therefore is also related to higher anti-HPV 16 level. This was also found in Mongolia study (115). For anti-HPVs, there is no relationship to age at first menstruation or first sexual intercourse, or contraceptive use, consistent with many studies worldwide (114;
115; 121; 130). One study showed that IUD usage was significantly related to persistent infection of HR-HPV (132). We checked HPV seropositivity, especially HPV 16 or 18, among women considering their use of hormonal contraceptives, condom, or IUD. There is no such correlation found in our study.
HPV seropositivity stratified by cytology to detect precancerous lesions confirmed that in women with HSIL lesions, anti-HPV 16 was significantly high with odds ratio of 3.69, adjusted by DNA positivity, comparable to that detected with all seven oncogenic types in Mongolia study (115). Such significant HPV 16 antibody presence was not observed in women with ASCUS or LSIL. In contrast, HPV 18 seropositivity in women with any abnormal cervical cytology was comparable to those with normal cytology. The difference of seropositivity of these two most oncogenic types between women of normal or abnormal cervical status suggests that HPV 16 seropositivity could possibly be an indicator of cervical lesion progress. However, low prevalence of anti-
HPV 18 in China might be another reason of insignificant relationship of HPV 18
101 seropositivity to high-grade cervical lesion. Due to relatively large sample size of our
study, this result is compelling despite this pattern was not found in other study (121).
The same case applied when using pathology, the “golden standard”, to detect cervical
cancer or precancerous lesions. Serological response to HPV 16 antigen was greatly
increased in women with CIN2 or above but not in women with CIN 1. A similar odds
ratio was found for both HPV 16 and 18 in a large population-based study in Costa Rica
(130). However, in our study we did not find such risk for anti-HPV 18, possibly due to smaller sample size of HPV 18 positive patients. The association between cumulative expose to HPV and cervical cancer development could be evidenced by high HPV seropositivity in women with abnormal cervical lesions. However, neither of anti-HPV
16 or 18 was significantly high in women with abnormal cervix using VIA as screening method, which is different from using above two methods. The relatively low sensitivity and specificity of VIA method might explain the situation. Thereafter only overall
seropositivity to any of four HPV types was slightly high in women with high-grade
cervical lesions detected by VIA method.
Low concordance was found when comparing HC2 positivity and seropositivity,
which was in an understandable range since only a subset of women seroconvert and
HC2 test is not a genotyping assay, either. Also, while HC2 measures positivity of 13
high risk types and 5 low risk types, our serology test detected only four types.
Concordance level was increased between DNA results from Linear Array and serology test, due to ability to genotype and higher sensitivity of Linear Array (data not shown).
Still less than half of women with positive DNA had antibody of corresponding type detected. Since Linear Array test was only preformed on all but five HC2 positive
102 samples and two hundred more HC2 negative samples randomly selected, but not on all
4211 samples, comparison with serology results weakened its meaning in contrast to a complete study on concordance between DNA and antibody tests. Therefore, DNA positivity in Linear Array test was based on the following assumptions: (1) women tested by Linear Array had type specific results accordingly; (2) women with negative cervical
high risk and low risk HC2 and not tested by Linear Array had negative Linear Array
results; (3) women with positive cervical high risk or low risk HC2 and not tested by
Linear Array had missing Linear Array results. Because HPV DNA and antibody
positivity represent outcomes from either recent or past exposure (126), poor or
intermediate agreement is as expected and shown in other studies (114; 115; 121; 131).
The strength of our study is a multi-center study with large sample size. We covered five sites across China, including both rural and urban area. It gave us a good
estimate that HPV seroprevalence is intermediate in China. We also tried to include
women in a wide age range from 15 to 54 and stratified each five-year age group with
even number of women. The number of virgins we recruited is relatively large, providing
a sound basis for future planning of vaccination program. Also, both DNA and serology test used in this study are well validated and standardized. However, there are some limits in our study. Because seroprevalence among many different geographical areas in China might vary greatly, as observed on HPV DNA prevalence, data from five sites could not represent at National level. Also, women were recruited through posters or advertisement,
but not by random sampling, indicating possible selection bias in our study. Self-reported
sexual behavior by female subjects might introduce some bias, too, when evaluating
association with sexual activity or seroprevalence among those who claimed to never
103 have sexual activity. Also, interviewer effect might exist because all subjects were not
interviewed by the same interviewer.
In summary, HPV seroprevalence is relatively low in China but it may
underestimate lifetime exposure in women. There is constant exposure to HPV infection,
particular HPV 6, among women at age of 20 or above and the chance of exposure is
increasing with age. Multiple-valent vaccination should be given to young generation not
exposing to HPV infection, such as before 20 years old.
4.5 ACKNOWLEDGEMENTS
I sincerely thank Dr. Youlin Qiao for insightful comments and helpful suggestions on this chapter. This work was done as I did one-year Fogarty International Clinical
Research Scholarship program in Department of Cancer Epidemiology, Cancer
Institute/Hospital, Chinese Academy of Medical Sciences, in Beijing, China. This scholarship was sponsored by Fogarty International Center, National Institute of Health.
This work was conducted under the supervision of Dr. Qiao at CICAMS and incorporated into this dissertation with the endorsement of my Ph.D. advisor, Dr. Au, at
OSU.
104 18 Sexually Active Women 16 Virgins 14
12 Total
10
8
6
4
Seropositivity(%) within group 2
0 HPV 6 HPV 16 HPV 11 HPV 18 HPV 16 or HPV 6 or HPV 18 11 6/11/16/18 HPV type
Figure 4.1 Seropositivity of anti-HPV 16, 18, 6, 11, 16 or 18, 6 or 11, and any of four types in sexually active women, virgins and overall women.
105 50 HPV 6 HPV 11 40 HPV 16
30 HPV 18 HPV 6/11/16/18 20 Seropositivity (%) Seropositivity 10
0 15-24 25-29 30-34 35-39 40-44 45-49 50-54 Age (Yrs)
Figure 4.2 Age –specific HPV 6, 11, 16 and/or 18 seroprevalence rates in Chinese women.
106 Anti-HPV 6/11/16/18
Variable Categories n % OR* (95% CI) Basic sociodemographic characteristics Age (years) 15-24 484 13.02 1.0 25-34 1241 14.50 1.19(0.87-1.63) 35-44 1252 17.25 1.40(1.03-1.91) 45-54 1234 16.69 1.40(1.02-1.90) Race Han Ethic Group 3328 16.17 1.0 Uyghur Ethnic Group 883 14.38 0.99(0.80-1.23) Regions Metropolitan 1566 18.65 1.0 Rural 1762 13.96 0.72(0.60-0.87) Education Primary School or Lower 1352 14.13 1.0 level (0-6 year) Secondary School or 2851 16.49 1.14(0.95-1.38) Higher (>= 7 year) Marital Single 34 26.47 1.0 status Married 3990 15.24 0.50(0.22-1.12) Widowed/Separated/ 185 25.95 0.88(0.37-2.09) Divorced Lifestyle charactersitics Smoking Never Smoked 4078 15.64 1.0 Former/Current Smoker 131 19.85 1.19(0.76-1.86) Drinking Never 3204 15.82 1.0 Ever 832 16.47 1.00(0.81-1.23)
Continued
Table 4.1 Odds ratios (ORs) and 95% confidence intervals (CIs) for seropositivity according to sociodemographic and related variables. *OR: adjusted for age and HPV
DNA positivity.
107 Table 4.1 Continued
Anti-HPV 6 Anti-HPV 16
Variable Categories n OR* (95% CI) Basic sociodemographic characteristics Age (years) 15-24 484 1.0 1.0 25-34 1241 1.03(0.69-1.53) 1.59(0.96-2.62) 35-44 1252 1.10(0.75-1.63) 1.64(1.00-2.69) 45-54 1234 1.06(0.72-1.58) 1.59(0.97-2.63) Race Han Ethic Group 3328 1.0 1.0 Uyghur Ethnic Group 883 1.40(1.08-1.83) 0.73(0.51-1.04) Regions Metropolitan 1566 1.0 1.0 Rural 1762 0.97(0.75-1.25) 0.79(0.60-1.04) Education Primary School or 1352 1.0 1.0 level Lower (0-6 year) Secondary School or 2851 0.81(0.64-1.03) 1.53(1.13-2.06) Higher (>= 7 year) Marital Single 34 1.0 1.0 status Married 3990 0.96(0.29-3.21) 0.76(0.22-2.63) Widowed/Separated/ 185 1.96(0.55-7.00) 1.09(0.29-4.13) Divorced Lifestyle charactersitics Smoking Never Smoked 4078 1.0 1.0 Former/Current 131 1.53(0.89-2.63) Smoker 0.82(0.39-1.71) Drinking Never 3204 1.0 1.0 Ever 832 1.04(0.79-1.37) 0.98(0.72-1.34)
Continued
108 Table 4.1 Continued
Anti-HPV 6/11/16/18 Variable Categories n % OR* (95% CI) Menstrual and reproductive characteristics Age at first <=13 1042 14.88 1.0 menstruation 14-16 246815.80 1.10(0.90-1.36) >=17 697 17.22 1.14(0.87-1.50) Number of 0 155 14.19 1.0 Pregnancies 1 785 15.03 1.03(0.62-1.70) 2 119615.55 0.97(0.59-1.59) 3 or more 2051 16.33 1.02(0.62-1.66) Abortion spontaneous only 356 13.48 1.0 voluntary only 187917.83 1.39(0.99-1.93) both 271 15.87 1.23(0.78-1.93) Sexual characteristics Age at first sexual <20 1249 14.73 1.0 intercourse 20-24 211215.81 1.02(0.83-1.24) >=25 844 17.30 1.13(0.89-1.45) Ever used contraceptive No 435 17.24 1.0 measures Yes 371115.71 0.85(0.65-1.12)
Continued
109 Table 4.1 Continued
Anti-HPV 6 Anti-HPV 16 Variable Categories n OR* (95% CI) Menstrual and reproductive characteristics Age at first <=13 1042 1.0 1.0 menstruation 14-16 24681.00(0.76-1.31) 1.23(0.89-1.68) >=17 697 1.09(0.76-1.56) 1.24(0.82-1.86) Number of 0 155 1.0 1.0 Pregnancies 1 785 0.73(0.40-1.33) 1.50(0.62-3.62) 2 11960.70(0.39-1.27) 1.59(0.67-3.80) 3 or more 2051 0.85(0.48-1.53) 1.58(0.67-3.76) spontaneous Abortion 356 1.0 1.0 only voluntary 1879 1.17(0.75-1.82) 1.74(1.03-2.93) only both 271 1.43(0.80-2.54) 0.94(0.44-2.03) Sexual characteristics Age at first sexual <20 1249 1.0 1.0 intercourse 20-24 2112 0.89(0.69-1.15) 1.17(0.86-1.60) >=25 844 0.69(0.49-0.98) 1.30(0.89-1.90) Ever used contraceptive No 435 1.0 1.0 measures Yes 3711 0.71(0.51-0.99) 0.86(0.58-1.29)
Continued
.
110 Table 4.1 Continued
Anti-HPV 6/11/16/18
Variable Categories n % OR* (95% CI) Sexual characteristics Lifetime no. sex 1 3230 13.50 1.0 partners 2 634 23.34 1.88(1.52-2.33) 3 194 21.13 1.54(1.07-2.23) >=4 123 28.46 2.34(1.54-3.54) Husband's extramarital No 2044 13.36 1.0 sexual relationships Yes 856 22.66 1.80(1.46-2.22) Unknown 152615.53 1.15(0.95-1.39)
Continued
111 Table 4.1 Continued
Anti-HPV 6 Anti-HPV 16
Variable Categories n OR* (95% CI) Sexual characteristics Lifetime no. sex 1 3230 1.0 1.0 partners 2 634 1.77(1.34-2.35) 2.43(1.81-3.28) 3 194 1.85(1.17-2.90) 1.40(0.80-2.46) >=4 123 2.12(1.24-3.62) 2.21(1.22-4.01) Husband's extramarital No 2044 1.0 1.0 sexual relationships Yes 856 1.91(1.46-2.50) 1.81(1.32-2.48) Unknown 1526 1.11(0.86-1.44) 1.39(1.04-1.86)
112
Diagnosis Lesion OR* (95% CI) Methods Anti-HPV Anti-HPV 16 Anti-HPV 18 6/11/16/18 Cytology Normal 1.0 1.0 1.0 ASC-US 1.29(0.96-1.73) 1.20(0.78-1.83) 1.08(0.53-2.19) LSIL 1.10(0.64-1.89) 0.90(0.42-1.91) 0.93(0.27-3.23) ≥HSIL 3.69(2.12-6.43) 5.12(2.86-9.19) 0.67(0.15-2.97)
Pathology Normal 1.0 1.0 1.0 CIN1 1.02(0.65-1.60) 1.19(0.66-2.16) 0.17(0.02-1.24) ≥CIN2 3.74(2.21-6.33) 5.89(3.39-10.23) 0.75(0.22-2.58)
VIA Normal 1.0 1.0 1.0 Low
grade 0.95(0.74-1.22) 0.99(0.68-1.43) 0.60(0.29-1.26) ≥High
grade 1.74(1.06-2.85) 1.81(0.96-3.41) 1.16(0.35-3.81)
Table 4.2 Anti-HPV 16, 18 or 6/11/16/18 serological response and diagnostic approach to define cervical intraepithelial lesions. *OR: adjusted for age and HPV DNA positivity.
113
n % OR (95% CI) HC2 Anti-HPV 16/18 High risk Negative 3687 5.94 1.0 Positive 524 21.18 4.26(3.31-5.47)
Anti-HPV 6/11 Low risk Negative 4093 9.65 1.0 Positive 118 16.95 1.91(1.17-3.12)
Anti-HPV 16/18 and 6/11 Both Negative 3611 1.47 1.0 Positive 42 7.14 5.16(1.55-17.23)
Table 4.3 Comparison of HPV seropositivity and Hybrid Capture 2 result.
114
Linear Array n % OR (95% CI） CIN1 Anti-HPV 16 HPV 16 Negative 94 7.45 1.0 Positive 19 47.37 11.19(3.46-36.99) Anti-HPV 18 HPV 18 Negative 106 0.00 1.0 Positive 7 14.29 N/A Anti-HPV 6 HPV 6 Negative 113 12.39 1.0 Positive 0 0.00 N/A Anti-HPV 11 HPV 11 Negative 110 5.45 1.0 Positive 3 66.67 34.67(2.77-442.67) >=CIN2 Anti-HPV 16 HPV 16 Negative 21 23.81 1.0 Positive 47 57.45 4.32(1.45-14.53) Anti-HPV 18 HPV 18 Negative 62 3.23 1.0 Positive 6 16.67 6.00(0.47-79.52) Anti-HPV 6 HPV 6 Negative 68 14.71 1.0 Positive 0 0.00 N/A Anti-HPV 11 HPV 11 Negative 66 6.06 1.0 Positive 2 50.00 15.50(0.82-301.01)
Table 4.4 Comparison of HPV seropositivity and Linear Array result.
115
CHAPTER 5
ATTRIBUTABLE CAUSES OF BREAST CANCER AND OVARIAN CANCER
TO REPRODUCTIVE FACTORS AND ORAL CONTRACEPTIVES IN CHINA
5.1 INTRODUCTION
Both genetic and environmental factors have been found to be involved in cancer development and occurrence. Breast cancer and ovarian cancer, two leading cancers in women, are linked to a number of environmental factors that influence female reproductive system and its function, except genetic change of oncogene and/or tumor suppressor gene (133; 134). Family planning policy implemented in China from late 70s and lifestyle change from economic improvement in recent three decades in this country has possibly changed reproductive factors (RFs). We used nulliparity, parity, age at first birth and duration of breastfeeding to study their attributable fraction (AF) in breast
cancer. Increased nulliparity, reduced multiparity, late age at first birth, and shorter
period of breastfeeding are associated with higher risk of breast cancer (133). Risk factor
of ovarian cancer involves reproductive component as well (134). We selected parity and
looked at contribution from its change to ovarian cancer.
Due to similar chemicals to hormone replacement therapy (HRT), oral
contraceptives (OC) use contributes to modest risk of breast cancer (66), while it has
116 clearly been shown to reduce the risk of ovarian cancer (134). The distribution of
contraception methods in many countries may be very different. Considering family planning policy in China, OC use is only one of contraception methods used among women. In this chapter we evaluated the attributable fraction of breast cancer due to OC use in each five-year age group among Chinese women.
Based on different situation and attitude to OC use and potential change of AFs over time in China, we presented prevalence of these factors, calculated their attributable
fractions to breast cancer and ovarian cancer, and evaluated attributable risk of these
factors in both cancer incidence and mortality.
5.2 MATERIALS AND METHODS
5.2.1 Data used for relative risk estimates of reproductive factors
We conducted systematic publication search on pubmed or cited references
regarding risk factors of breast cancer and ovarian cancer. Language was limited to
English and Chinese. Search terms included risk factor, reproductive factor, China, breast
cancer, ovarian cancer and social science. Meta-analysis was preferred over single
studies. Finally five meta-analyses based on China data in various regions were carefully reviewed. Nevertheless, these analyses either used controls not applicable to our data or had unclear definition on reproductive factors. Therefore, we decided to adopt relative risk (RR) estimates used by IARC in French study (54).
5.2.2 Data used for exposure prevalence of reproductive factors
To collect population-based data on prevalence of reproductive factors, yearbooks
(135-141) on health, population, family planning, fertility and reproductive health in
117 China from 1980 to 2008 were browsed. Meanwhile, publications or data analysis book
(142-151) from related random-sampling national surveys or investigation conducted
during this period were located. Department of Health or National Population and Family
Planning Commission performed these surveys or investigations in 1985, 1987, 1988,
1992, 1995, 1997, 2001, and 2004. Two dataset books (152; 153) were also used to include data from national population and fertility survey conducted in 1982 and 1990.
Therefore, we focused on yearbooks in these particular years to supplement data from surveys. We also looked at some papers (154; 155) presenting prevalence of reproductive
factors, though they are at local level.
Totally we were able to extract prevalence data of all five factors in 1982, 1987,
1988, 1990, 1992, 1995, 1997 and 2001. Sporadic data in other years were found for one or more of all factors, but not for all of them. In order to obtain maximal time interval, we used prevalence in 1982 and 2001 to calculate AFs of reproductive factors.
(1) For the prevalence of nulliparous women, 2001 survey (151) only provided percentage of women who never had children ever born among women at age of 40 to 49.
Accordingly, prevalence in 1982 (153) was calculated in the same age group to be comparable.
(2) For parity, the ratio of 1982 data (153) was total number of children ever born by women of 15-49 age divided by total number of parous women or women of this age group. In 2001 data (151), the numerator was calculated by addition of products, each of which was by multiplying number of children ever born with number of women who had this number of children ever born. The ratio of 2001 data was this numerator divided by total number of parous women or women of 15-49 ages. Again, 1982 prevalence covered
118 women from 15 to 49 to accommodate 2001 data, though it provided data of women from
15 to 64.
(3) For percentage of women with age at first birth equal to or greater than
30, overall distribution of married women from 15-49 by age at first birth in 2000 was
given (151) and used as surrogate of 2001 data. In 1982 (153), such distribution was
given at each age at survey, instead of an overall distribution. So we summed up all
numerators at each age point from 15 to 49 and all denominators, respectively, and then
took the percentage.
(4) For number of months breastfeeding, because of rare data on this factor, we
only found distribution of live births by months of pure breastfeeding on a five-year
range. So we calculated average number from 1980 to 1984 (145) to use as prevalence
estimate in 1982. The numerator was calculated by addition of products, each of which
was by multiplying number of months of pure breastfeeding with number of live births
who received this number of months of pure breastfeeding. Only total number of live
births that received more than eight months of pure breastfeeding was given. So we
multiplied this total number with 15 months of breastfeeding, which is a rough estimate
of mean number of months of breastfeeding longer than eight months. Then the ratio was
this numerator divided by total number of live births. The prevalence in 2001 was
calculated the same way, with smaller sample size and using 2000 data (151) as surrogates.
5.2.3 Data used for RR estimates of oral contraceptives
Similar systematic search for reproductive factors was conducted for oral contraceptives. Key words were oral contraceptives, breast cancer, China, and risk factor.
119 Four meta-analyses (156-159) were included to estimate RR of using oral contraceptives
over not using. The odds ratios from these four papers were quite consistent and
geometric mean was taken to use as RR estimate (1.46).
5.2.4 Data used for exposure prevalence of oral contraceptives
When extracting prevalence data for reproductive factors, age-specific prevalence
of oral contraceptives usage was found only in 2001 national survey (151). So we used
this set of data to estimate the situation in 2005, assuming there is no big difference on
age-specific prevalence among recent years. The sample size was 32464.
5.2.5 Cancer incidence and mortality data
Data of the Third National Death Cause Survey in China in 2004-2005 was used
to calculate AF of breast cancer and ovarian cancer to OC and RFs (Table 5.1). This
retrospective survey was conducted in 160 randomized counties and 53 high risk areas between 2004 and 2005. Cancer incidence data were estimated using Mortality and
Incidence (M/I) ratio and cancer deaths. The details of this survey and estimation
calculation were seen in [Ref].
5.2.6 AF calculation
AF is defined as a proportion of cancers in the total population that is attributed to
a risk factor. AF can be calculated by the following formula that was described by Levin
(60).
P *()RR −1 AF = [ P *()RR −1 ]+1
RR is relative risk; P is prevalence of exposure to a risk factor.
120 For continuous variables in risk factors, AF is obtained by multiplying the RR for unit exposure (RRu, e.g., RR for 1 g of alcohol/day) and the average exposure level (d), shown in the following formula.
Log RR = log (RRu) x d
AF = (RR-1)/RR
RRu is relative risk increase to unit exposure; d is average exposure level.
5.3 RESULTS
5.3.1 Exposure prevalence of reproductive factors
Exposure data of four reproductive factors in 1982 and 2001 were given in table
5.2. Percentage of women who never had child among all women aged 40 or above was merely changed over 20 years. Mean number of children per woman or parous woman significantly dropped from 2.04 and 3.28, to 1.44 and 1.80, respectively. Meanwhile, percentage of women who had first birth at age of 30 or older was increased three-fold in
2001, compared with in early 80s. Number of months that women breastfed their children also shortened by about 30%.
5.3.2 Attributable fraction change of breast cancer and ovarian cancer to
reproductive factor change
The corresponding AF was calculated in table 5.3 for each time point and the difference between AFs was obtained for each factor. The total change in AF due to reproductive factor change for breast cancer is 16.21%. We did sensitivity analysis to calculate AF change, using counterfactual exposure distribution to compare with data in
2001. Counterfactual exposure distribution assumes zero nulliparity, none of women
121 giving first birth after 30, average 2.2 children per woman, and 18 months of breastfeeding. The total change in AF for breast cancer is 11.36% (data not shown), which is in an acceptable range. The total AF change in ovarian cancer by change of four reproductive factors is 10.65%. The same sensitivity analysis using counterfactual exposure distribution gave 13.64% of AF change (data not shown).
5.3.3 Number of cancer cases and deaths attributable to reproductive
factor change
Table 5.4 calculated changes in breast and ovarian cancer incidence and mortality owing to changes in reproductive factors over time. There were 7394 breast cancer cases and 2129 breast cancer deaths attributable to change in reproductive factors, respectively, which accounts for 5% in both breast cancer incidence and mortality and less than 1% in all cancers.
5.3.4 Exposure prevalence of oral contraceptives
We presented age-specific prevalence of current OC use in Figure 5.1 and table
5.5. Except higher percentage of current OC users in very young women aged from 15 to
19, this percentage is constantly low around 2% in women more than 20 years old.
5.3.5 Attributable fraction and number of cases and deaths of breast cancer
attributable to oral contraceptives use
We presented age-specific AF of current OC use and cancer deaths and incidence attributable to OC use in different age groups in table 5.5. More women at age of 15-19 use OC for contraception, whereas only about 2% in women after 20 years old rely on
OC for birth control. This is understandable that older women tend to have tubal ligation or use IUD as major contraception approach. But younger women, especially before their
122 having child, choose to use OC. The overall prevalence of OC use in women aged 15-49 years old is 1.74% and the total AF change in breast cancer attributable to OC use is
0.80%. Considering low breast cancer cases and deaths in young age group, total numbers of breast cancer cases and deaths attributable to OC use in parous women are
347 and 100, respectively.
5.4 DISCUSSION
Although cancer incidence and mortality data in 2005 was used in our attributable risk study, we used AF change due to reproductive factor change estimated between 1982 and 2001 to make calculation of changes in cancer incidence and mortality. This is based on assumption that reproductive factor change is consistent over time. One big concern about exposure data of reproductive factors is that the big difference between sample sizes of data collected in 1982 and 2001. In 1982, a large survey about fertility and reproduction was taken to cover 60% (297274448/488636422) of Chinese women at that time. About 297 million women were surveyed to investigate prevalence of several important reproductive factors including parity data in China. However, national survey on family planning and reproduction health in 2001 only sampled 39586 women, which has ten thousand times difference from that in 1982, despite growing population in 2001.
Thus it is noteworthy that the big difference in sample size may infer incomparability in data.
For nulliparity data, except concern on sample size difference (44847298 in 1982 vs. 10435 in 2001), women aged from 40 to 49 years old only was considered since 2001 survey did not give data beyond age of 49. Small age range investigated might be one
123 reason that percentage of nulliparous women did not change much over two decades. It indicates that although there are more women receiving “modern” education or bearing such thoughts in recent decade in China, particularly in urban areas, most of Chinese women are still acceptable to traditional value about family and reproduction. Though in our study attributable cancer cases or deaths to nulliparity are negligible, more complete data over time is needed.
Family planning policy started in China in late 1970s or early 1980s. We collected data on mean number of children per parous woman or woman from 1982 to 2004.
Following the effectiveness of this policy, though the percentage of nulliparous women is quite constant over years, it showed the decreasing trend on number of children averaged on each woman. Therefore, mean number of children born by each parous woman dropped significantly in twenty years, which contributes to the greatest change in AF for breast cancer. Correspondingly, AF change due to decreasing mean number of children per woman for ovarian cancer also accounted for 10%. These changes in AF from parity change and consequent number of cancer cases and deaths attributable were more significant than in France (54), considering different social environment occurring in these two countries over time. Furthermore, because smaller sample size was used in
2001 survey (39586 in 2001 vs. 248036697 in 1982), and data in recent few years showed even lower mean number of children per parous woman or woman, change in AF due to this factor might be underestimated.
The effect of family planning policy on reproduction in China was also seen in regards to age at first birth. Recommendations on late marriage and birth from government have encouraged more women to choose to have first child at their own
124 willingness. Taking into account of enhanced education and career development, more and more women, especially from urban area, begin to delay their child-bearing plan to late phase. Percentage of age at first birth greater than or equal to 30 years old is expectedly increased over years. AF change due to this factor is, though, not as significant as due to parity for breast cancer. Less number of cancer cases and deaths was counted toward age at first birth, compared with French study (54). However, comparison of 2001 data in small sample size (1008 women) against 1982 data surveyed among
191629 women might give conservative estimates, since other sources data around 2005 suggested bigger proportion of women having their first birth after 30 years old.
We had most difficulty in locating data on duration of breastfeeding in our literature search. In accordance with delayed age at first birth among Chinese women, number of months breastfeeding was also reduced, which is not the case in France (54).
On the other hand, as we used 15 months to roughly estimate mean number of months of pure breastfeeding longer than 8 months (see section 5.2.2), it is possible that duration of breastfeeding is conservatively estimated at each time point. Since sample sizes are comparable in two surveys (1569 in 2001 vs. 4702 in 1982), estimation of this factor is not influenced by sample size difference as for other factors. Still, more studies should be conducted to give more accurate estimates on duration of breastfeeding.
In overall, there was 16.2% increase in fraction of breast cancer attributable to these four aspects of reproductive factors, mostly from decreased number of children per parous woman, and 10.7% increase in fraction of ovarian cancer from similar parity factor. It suggests in the process of implementing family planning policy, health consideration should be taken into account and related health program might be given for
125 disease prevention and control. It is worth noting that those estimates are limited to women aged 15 to 49 years old, but most breast cancer cases occur after 40 years old, but not among women at parous age. The same applied to ovarian cancer. Though AF increase is smaller in France and contribution from four factors varies between China and
France (54), overall percentage of breast cases or deaths attributable to change in reproductive factors was comparable in two countries and stayed around 5%. Since there is lack of reproductive data among women in wider age range, this percentage might be underestimated and big surveys covering older women to represent reproduction status for the whole population are in great need.
Age-stratified percentage of OC use was extracted from 2001 survey to estimate situation in 2005. The same assumption on prevalence constancy is applied since no lag time is considered for OC use. Some studies (66) found that OC use contributes different risk of breast cancer in younger and older women. But most of their studies used 40 or 45 years old as boundary to separate younger and older age groups, while our data surveyed women from 15 to 49 years old. Therefore, AF calculation in our study assumes the same risk for different age groups. Overall AF increase in breast cancer due to OC use is very low and so are cancer cases and deaths attributable to OC use. This is largely because of very low prevalence of OC use in China.
Kelsey et al. reviewed that many reproductive factors, such as early age at menarche or late age at menopause, are identified as risk factors of breast cancer (133) and tubal ligation is protective against ovarian cancer (134). These reproductive factors were not studied in this paper, but they might have effect on cancer incidence and mortality as well. Also the time interval we chose is relatively short since long-term
126 change on cancer incidence and mortality due to reproductive factor change is more obvious. Shortage of data, especially before 1980, makes prolonged interval unattainable.
Nevertheless, the merit of our study is that we used national data instead of local data in
AF calculation, which could be applied to general population. As the first study on attributable fraction of breast cancer and ovarian cancer to reproductive factors and oral contraceptives in China, our report provided estimation of attributable number and percentage of cancer cases and deaths as well as a guide of direction-to-go in cancer prevention and control.
5.5 ACKNOWLEDGEMENTS
I sincerely thank Dr. Youlin Qiao for insightful comments and helpful suggestions on this chapter, Jianbing Wang for cancer incidence and mortality data reference. This work was done as I did one-year Fogarty International Clinical Research Scholarship program in Department of Cancer Epidemiology, Cancer Institute/Hospital, Chinese
Academy of Medical Sciences (CICAMS), in Beijing, China. This scholarship was sponsored by Fogarty International Center, National Institute of Health.
This work was conducted under the supervision of Dr. Qiao at CICAMS and incorporated into this dissertation with the endorsement of my Ph.D. advisor, Dr. Au, at
OSU.
127
Women Site (age group) Deaths Cases
Breast （ 40-49） 8961 31325 Breast （ 15-49） 13110 45529 Breast （ 15-19） 9 26 Breast （ 20-24） 69 211 Breast （ 25-29） 262 917 Breast （ 30-34） 845 2945 Breast （ 35-39） 2964 10097 Breast （ 40-44） 4370 15348 Breast （ 45-49） 4591 15977 Breast （ 40-65） 27153 94289 Breast (all age 40134 142732 groups) Ovaries (15-49) 2377 5751
Ovaries (40-65) 5907 14310 Ovaries (all age 10482 25616 groups)
Table 5.1 Number of Deaths and Cases of Breast and Ovarian Cancer in 2005 in China.
128
Exposure Exposure Reproductive factor RR* in 1982 in 2001
% Nulliparous 1.5% 1.2% 1.36
Mean Number of Children per Parous Woman (for breast cancer) 3.28 1.80 0.93 risk reduction per child Mean Number of Children per Woman (for ovarian cancer) 2.04 1.44 0.87 risk reduction per child
% with age at first birth >= 30 years 1.3% 4.9%# 1.67
risk reduction per 12 Number of Months Breastfeeding 9.4† 6.5¶ 0.957 months of breastfeeding Total change in AF for breast cancer
Table 5.2 Change in reproductive factors between 1982 and 2001 in China.
* RR adopted from attributable fraction study in France by IARC (22) # Exposure in 2001 on % with age at first birth >= 30 years used data in 2000. † Number of month’s breastfeeding in 1982 was average data estimated from 1980 to 1984. ¶ Number of months breastfeeding in 2001 used data in 2000
129
Difference Reproductive factor AF 1982 AF 2001 in AF‡
% Nulliparous 0.54% 0.43% -0.11%
Mean Number of Children per Parous Woman (for breast cancer) -26.87% -13.95% 12.92% Mean Number of Children per Woman (for ovarian cancer) -32.86% -22.21% 10.65%
% with age at first birth >= 30 years 0.86% 3.18% 2.32%
Number of Months Breastfeeding -3.50% -2.42% 1.08% Total change in AF for breast cancer 16.21%
Table 5.3 Change in AF between 1982 and 2001 in China.
‡ AF calculation for nulliparity and % with age at first birth >= 30 years was based on ordered RRs. AF calculation for number of children and number of months breastfeeding was based on continuous RRs.
130 INCIDENCE Females No. Cancer N AF Attributable Ovary - Number of children Ovary 5751 10.65% 612 Breast >= 35 Breast - Nulliparity years 31325 -0.11% -34 Breast - Number of Breast among children parous women 45529 12.92% 5883 Breast - Breast among Breastfeeding parous women 45529 1.08% 491 Breast - Age at first Breast among birth parous women 45529 2.32% 1054 Breast cancer cases attributable to change in reproductive factors 7394 Breast cancer % 5.18% All cancers Total 8007 % 0.76% MORTALITY Females No. Cancer N AF Attributable Ovary - Number of children Ovary 2377 10.65% 253
Breast >= 35 Breast - Nulliparity years 8961 -0.11% -10 Breast - Number of Breast among children parous women 13110 12.92% 1694 Breast - Breast among Breastfeeding parous women 13110 1.08% 141 Breast - Age at first Breast among birth parous women 13110 2.32% 304 Breast cancer cases attributable to change in reproductive factors 2129 Breast cancer % 5.31% All cancers Total 2382 % 0.36%
Table 5.4 Estimation of the number of breast and ovarian cancers cases and deaths in
China in 2005 attributable to changes in reproductive risk factors between 1982 and 2001.
131
All All No. breast No. breast % Current breast breast cancer cases cancer deaths Age AF* OC use cancer cancer attributable attributable to cases deaths to OC use OC use 15-19 8.82% 3.90% 26 9 1 0 20-24 2.18% 0.99% 211 69 2 1 25-29 1.80% 0.82% 917 262 8 2 30-34 1.76% 0.80% 2945 845 24 7 35-39 1.62% 0.74% 10097 2964 75 22 40-44 2.11% 0.96% 15348 4370 148 42 45-49 1.23% 0.56% 15977 4591 90 26 BCs 15-49 45521 13110 347 100 % 0.76% 0.76% All BCs 142732 40134 % All BCs 0.24% 0.25% % All Cancers 0.03% 0.02%
Table 5.5 Prevalence of current OC use in women 15–49 years old in China and attribute numbers of breast cancer (BC) cases and deaths.
* AF calculation takes RR value of 1.46.
132 10
8
6
4 users (%)
2
Pervalence of current OC 0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 Age (Yrs)
Figure 5.1 Age-stratified prevalence of current OC users in Chinese women (2001 data).
133
CHAPTER 6
PERSPECTIVES AND CONCLUSION
For the purpose to identify dose range of investigational agent that achieved the desired effectiveness, selection of PD endpoint or biomarker under preclinical conditions is being incorporated into clinical experience, aiming to screen for promising agents in early clinical exploration (160). One half of the research in this dissertation was to evaluate γH2AX, a marker of DNA damage, as a PD endpoint of tumor sensitivity to combinations of suramin treatment to DNA-damaging agent. The other half of this dissertation focused on clinical experience in cancer epidemiology. Prevalence of exposure to leading risk factor of cervical cancer and attributable causes of breast cancer and ovarian cancer in Chinese women were first reported.
Research of this dissertation stated that the severity of DNA damage, measured by peak level and repair kinetics of γH2AX, was correlated with the cytotoxicity to cisplatin, and that addition of low dose suramin did not increase the extent of DSB but sustained the DSB. On the contrary, low dose suramin appeared not to alter the DSB induced by taxanes under both in vitro and in vivo conditions. These results provide us clue on mechanism of suramin sensitization to cisplatin through biochemical pathways in DNA damage response. On the other hand, our results conferred the possibility that suramin
134 sensitization effect observed in multiple experimental models is due to specific mechanism dependant on treatment type. Therefore, whether marker of DNA damage can be used as PD endpoint in other DNA-damaging treatment should be investigated carefully between chemotherapeutics. Clinical application of our preclinical findings is potentially possible, based on a recent study that developed and validated an immunocytochemical assay to quantitatively measure γH2AX in human blood and tumor biopsies (161). This study supports γH2AX as a robust PD biomarker with more potential than other markers of DNA damage in clinical monitoring of DNA damage in human samples. More translational research is necessary to explore the usage of γH2AX as PD endpoint of suramin sensitization in preclinical model and extend to clinical practice.
Research on the cancer epidemiology studies described the population-based sero- prevalence of HPV 6, 11, 16, 18, leading causes of cervical cancer and genital warts, in
Chinese women. The overall and age-stratified prevalence of HPV antibodies is not high in China, compared with other countries worldwide. Sexual activity is most influential among factors associated with HPV antibody presence, a marker of lifetime exposure to
HPV. Antibody level against HPV is also correlated with severity of cervical lesions as it measures cumulative exposure to HPV. Considering the emerging extensive changes in population and lifestyle due to modernization in China, the burden of cervical cancer can not be ignored. Primary prevention using prophylactic HPV vaccines and secondary prevention with country-wide screening programs warrants serious consideration in
China.
We also presented attributable causes of reproductive factor change and oral contraceptives use to breast cancer and ovarian cancer in China. Modest fraction of breast
135 cancer and ovarian cancer attributable to reproductive factors change was calculated, mostly contributed by decreasing mean number of children per woman or parous woman.
Oral contraceptives use has insignificant influence on cancer incidence and mortality due to its less prevalence use as contraception method in China. These data provide basis for future policy makers in control and prevention of breast cancer and ovarian cancer in
Chinese women, offering early screening and treatment program to accompany family planning policy implemented in China.
With translational research emerging as promising field to connect preclinical discovery in the laboratory to clinical outcome efficiently, concept of phase “0” trial was proposed, which aims to integrate PD assays into early clinical investigations and facilitate rational selection of anti-cancer drugs (162). Our preclinical findings that
γH2AX as PD endpoint of suramin sensitization is in a great prospective to be translated into future biomarker testing during clinical application of chemotherapy combined with suramin. On the other hand, our clinical experience in cancer epidemiology renders a starting point for researcher in basic science and provokes more insightful thoughts to better cancer control and prevention.
136
BIBLIOGRAPHY
1. A. Jemal, R. Siegel, E. Ward, Y. Hao, J. Xu, T. Murray, and M. J. Thun. Cancer statistics, 2008, CA Cancer J. Clin., 58:71-96 (2008).
2. D. M. Parkin, F. Bray, J. Ferlay, and P. Pisani. Global cancer statistics, 2002, Ca-A Cancer Journal for Clinicians, 55:74-108 (2005).
3. A. Sancar, L. A. Lindsey-Boltz, K. Unsal-Kacmaz, and S. Linn. Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints, Annual Review of Biochemistry, 73:39-85 (2004).
4. X. H. Yang and L. Zou. Checkpoint and coordinated cellular responses to DNA damage, Results Probl. Cell Differ., 42:65-92 (2006).
5. W. K. Kaufmann and R. S. Paules. DNA damage and cell cycle checkpoints, FASEB J., 10:238-247 (1996).
6. J. Bartek and J. Lukas. Mammalian G1- and S-phase checkpoints in response to DNA damage, Curr. Opin. Cell Biol., 13:738-747 (2001).
7. J. Bartek, C. Lukas, and J. Lukas. Checking on DNA damage in S phase, Nat. Rev. Mol. Cell Biol., 5:792-804 (2004).
8. G. R. Stark and W. R. Taylor. Control of the G2/M transition, Mol. Biotechnol., 32:227-248 (2006).
9. A. Musacchio and E. D. Salmon. The spindle-assembly checkpoint in space and time, Nat. Rev. Mol. Cell Biol., 8:379-393 (2007).
10. L. M. Dejean, S. Martinez-Caballero, and K. W. Kinnally. Is MAC the knife that cuts cytochrome c from mitochondria during apoptosis?, Cell Death. Differ., 13:1387-1395 (2006).
11. I. Ben Porath and R. A. Weinberg. The signals and pathways activating cellular senescence, International Journal of Biochemistry & Cell Biology, 37:961-976 (2005).
12. G. P. Dimri. What has senescence got to do with cancer?, Cancer Cell, 7:505-512 (2005).
137 13. J. Kao, M. T. Milano, A. Javaheri, M. C. Garofalo, S. J. Chmura, R. R. Weichselbaum, and S. J. Kron. gamma-H2AX as a therapeutic target for improving the efficacy of radiation therapy, Current Cancer Drug Targets, 6:197-205 (2006).
14. J. P. Banath and P. L. Olive. Expression of phosphorylated histone H2AX as a surrogate of cell killing by drugs that create DNA double-strand breaks, Cancer Research, 63:4347-4350 (2003).
15. Y. H. Hsiang, M. G. Lihou, and L. F. Liu. Arrest of Replication Forks by Drug-Stabilized Topoisomerase I-Dna Cleavable Complexes As A Mechanism of Cell Killing by Camptothecin, Cancer Research, 49:5077-5082 (1989).
16. G. D. Wilson, S. M. Bentzen, and P. M. Harari. Biologic basis for combining drugs with radiation, Seminars in Radiation Oncology, 16:2-9 (2006).
17. X. Huang, M. Okafuji, F. Traganos, E. Luther, E. Holden, and Z. Darzynkiewicz. Assessment of histone H2AX phosphorylation induced by DNA topoisomerase I and II inhibitors topotecan and mitoxantrone and by the DNA cross- linking agent cisplatin, Cytometry Part A, 58A:99-110 (2004).
18. E. Mladenov, I. Tsaneva, and B. Anachkova. Activation of the S phase DNA damage checkpoint by mitomycin C, Journal of Cellular Physiology, 211:468-476 (2007).
19. B. Ewald, D. Sampath, and W. Plunkett. H2AX phosphorylation marks gemcitabine-induced stalled replication forks and their collapse upon S-phase checkpoint abrogation, Molecular Cancer Therapeutics, 6:1239-1248 (2007).
20. Y. M. Ma, H. H. Lu, K. Schwarz, and M. R. Lieber. Repair of double- strand DNA breaks by the human nonhomologous DNA end joining pathway - The iterative processing model, Cell Cycle, 4:1193-1200 (2005).
21. S. J. Collis, T. L. DeWeese, P. A. Jeggo, and A. R. Parker. The life and death of DNA-PK, Oncogene, 24:949-961 (2005).
22. E. J. Park, D. W. Chan, J. H. Park, M. A. Oettinger, and J. Kwon. DNA- PK is activated by nucleosomes and phosphorylates H2AX within the nucleosomes in an acetylation-dependent manner, Nucleic Acids Research, 31:6819-6827 (2003).
23. J. T. Reardon, A. Vaisman, S. G. Chaney, and A. Sancar. Efficient nucleotide excision repair of cisplatin, oxaliplatin, and bis-aceto-ammine-dichloro- cyclohexylamine-platinum(IV) (JM216) platinum intrastrand DNA diadducts, Cancer Research, 59:3968-3971 (1999).
24. A. Celeste, O. Fernandez-Capetillo, M. J. Kruhlak, D. R. Pilch, D. W. Staudt, A. Lee, R. F. Bonner, W. M. Bonner, and A. Nussenzweig. Histone H2AX
138 phosphorylation is dispensable for the initial recognition of DNA breaks, Nature Cell Biology, 5:675-U51 (2003).
25. A. Takahashi and T. Ohnishi. Does gamma H2AX foci formation depend on the presence of DNA double strand breaks?, Cancer Letters, 229:171-179 (2005).
26. M. Stucki and S. P. Jackson. gamma H2AX and MDC1: Anchoring the DNA-damage-response machinery to broken chromosomes, Dna Repair, 5:534-543 (2006).
27. D. Chowdhury, M. C. Keogh, H. Ishii, C. L. Peterson, S. Buratowski, and J. Lieberman. gamma-H2AX dephosphorylation by protein phosphatase 2A facilitates DNA double-strand break repair, Molecular Cell, 20:801-809 (2005).
28. C. R. Lu, F. Zhu, Y. Y. Cho, F. Q. Tang, T. Zykova, W. Y. Ma, A. M. Bode, and Z. G. Dong. Cell apoptosis: Requirement of H2AX in DNA ladder formation, but not for the activation of caspase-3, Molecular Cell, 23:121-132 (2006).
29. A. Rebbaa, X. Zheng, F. Chu, and B. L. Mirkin. The role of histone acetylation versus DNA damage in drug-induced senescence and apoptosis, Cell Death and Differentiation, 13:1960-1967 (2006).
30. E. P. Rogakou, W. Nieves-Neira, C. Boon, Y. Pommier, and W. M. Bonner. Initiation of DNA fragmentation during apoptosis induces phosphorylation of H2AX histone at serine 139, Journal of Biological Chemistry, 275:9390-9395 (2000).
31. F. D. di Fagagna, P. M. Reaper, L. Clay-Farrace, H. Fiegler, P. Carr, T. von Zglinicki, G. Saretzki, N. P. Carter, and S. P. Jackson. A DNA damage checkpoint response in telomere-initiated senescence, Nature, 426:194-198 (2003).
32. M. G. Hovest, N. Bruggenolte, K. S. Hosseini, T. Krieg, and G. Herrmann. Senescence of human fibroblasts after psoralen photoactivation is mediated by ATR kinase and persistent DNA damage foci at telomeres, Molecular Biology of the Cell, 17:1758-1767 (2006).
33. C. R. Middaugh, H. Mach, C. J. Burke, D. B. Volkin, J. M. Dabora, P. K. Tsai, M. W. Bruner, J. A. Ryan, and K. E. Marfia. Nature of the Interaction of Growth- Factors with Suramin, Biochemistry, 31:9016-9024 (1992).
34. F. Manetti, F. Corelli, and M. Botta. Fibroblast growth factors and their inhibitors, Current Pharmaceutical Design, 6:1897-1924 (2000).
35. C. A. Stein, T. M. Khan, Z. Khaled, and J. L. Tonkinson. Cell surface binding and cellular internalization properties of suramin, a novel antineoplastic agent, Clinical Cancer Research, 1:509-517 (1995).
139 36. D. F. Mccain, L. Wu, P. Nickel, M. U. Kassack, A. Kreimeyer, A. Gagliardi, D. C. Collins, and Z. Y. Zhang. Suramin derivatives as inhibitors and activators of protein-tyrosine phosphatases, Journal of Biological Chemistry, 279:14713- 14725 (2004).
37. J. Li, L. K. Li, J. F. Ma, L. H. Wei, M. Niyazi, C. Q. Li, A. D. Xu, J. B. Wang, H. Liang, J. Belinson, and Y. L. Qiao. Knowledge and attitudes about human papillomavirus (HPV) and HPV vaccines among women living in metropolitan and rural regions of China, Vaccine, 27:1210-1215 (2009).
38. B. S. Apgar, L. Zoschnick, and T. C. Wright. The 2001 Bethesda System terminology, American Family Physician, 68:1992-1998 (2003).
39. H. N. Nguyen and S. R. B. Nordqvist. The Bethesda system and evaluation of abnormal pap smears, Seminars in Surgical Oncology, 16:217-221 (1999).
40. M. Durst, L. Gissmann, H. Ikenberg, and H. Zurhausen. A Papillomavirus Dna from A Cervical-Carcinoma and Its Prevalence in Cancer Biopsy Samples from Different Geographic Regions, Proceedings of the National Academy of Sciences of the United States of America-Biological Sciences, 80:3812-3815 (1983).
41. V. Cogliano, R. Baan, K. Straif, Y. Grosse, B. Secretan, and F. El Ghissassi. Carcinogenicity of human papillomaviruses, Lancet Oncology, 6:204 (2005).
42. C. M. Wheeler. Natural History of Human Papillomavirus Infections, Cytologic and Histologic Abnormalities, and Cancer, Obstetrics and Gynecology Clinics of North America, 35:519-+ (2008).
43. A. Frega, M. Cenci, P. Stentella, L. Cipriano, A. De Ioris, M. Alderisio, and A. Vecchione. Human papillomavirus in virgins and behaviour at risk, Cancer Letters, 194:21-24 (2003).
44. H. zur Hausen. Papillomaviruses in the causation of human cancers - a brief historical account, Virology, 384:260-265 (2009).
45. M. Schiffman and P. E. Castle. Human papillomavirus - Epidemiology and public health, Archives of Pathology & Laboratory Medicine, 127:930-934 (2003).
46. J. Konya and J. Dillner. Immunity to oncogenic human papillomaviruses, Advances in Cancer Research, Vol 82, 82:205-238 (2001).
47. M. Arbyn, P. Sasieni, C. J. L. M. Meijer, C. Clavel, G. Koliopoulos, and J. Dillner. Clinical applications of HPV testing: A summary of meta-analyses, Vaccine, 24:78-89 (2006).
48. D. Opalka, C. E. Lachman, S. A. MacMullen, K. U. Jansen, J. F. Smith, N. Chirmule, and M. T. Esser. Simultaneous quantitation of antibodies to neutralizing
140 epitopes on virus-like particles for human papillomavirus types 6, 11, 16, and 18 by a multiplexed luminex assay 2, Clinical and Diagnostic Laboratory Immunology, 10:108- 115 (2003).
49. S. de Sanjose, M. Diaz, X. Castellsague, G. Clifford, L. Bruni, N. Munoz, and F. X. Bosch. Worldwide prevalence and genotype distribution of cervical human papillomavirus DNA in women with normal cytology: a meta-analysis, Lancet Infectious Diseases, 7:453-459 (2007).
50. F. X. Bosch, A. N. Burchell, M. Schiffman, A. R. Giuliano, S. de Sanjose, L. Bruni, G. Tortolero-Luna, S. K. Kjaer, and N. Munoz. Epidemiology and Natural History of Human Papillomavirus Infections and Type-Specific Implications in Cervical Neoplasia, Vaccine, 26:K1-K16 (2008).
51. J. S. Smith, A. Melendy, R. K. Rana, and J. M. Pimenta. Age-Specific Prevalence of Infection with Human Papillomavirus in Females: A Global Review, Journal of Adolescent Health, 43:S5-S25 (2008).
52. Ji, J. Age-specific Prevalence of Human Papillomavirus DNA and Antibody: Systematic Review. Chang, Y. and Smith, J. S. [13], 10. 2009. Abstract Book, The 25th International Papillomavirus Conference Clinical & Educational Workshop. Ji, J., Chang, Y., and Smith, J. S. Ref Type: Generic
53. P. Boffetta, J. K. McLaughlin, C. La Vecchia, P. Autier, and P. Boyle. 'Environment' in cancer causation and etiological fraction: limitations and ambiguities, Carcinogenesis, 28:913-915 (2007).
54. P. Boffetta, M. Tubiana, C. Hill, M. Boniol, A. Aurengo, R. Masse, A. J. Valleron, R. Monier, G. de The, P. Boyle, and P. Autier. The causes of cancer in France, Annals of Oncology, 20:550-555 (2009).
55. R. Doll and R. Peto. The Causes of Cancer - Quantitative Estimates of Avoidable Risks of Cancer in the United-States Today, Journal of the National Cancer Institute, 66:1191-& (1981).
56. C. L. Sistrom and C. W. Garvan. Proportions, odds, and risk, Radiology, 230:12-19 (2004).
57. Armitage P and Berry G. Statistical methods in medical research. 1987. London, Blackwell Scientific Publications. Ref Type: Generic
58. S. Wacholder, J. Benichou, E. F. Heineman, P. Hartge, and R. N. Hoover. Attributable Risk - Advantages of A Broad Definition of Exposure, American Journal of Epidemiology, 140:303-309 (1994).
141 59. C. J. L. Murray and A. D. Lopez. On the comparable quantification of health risks: Lessons from the Global Burden of Disease Study, Epidemiology, 10:594- 605 (1999).
60. M. L. LEVIN. The occurrence of lung cancer in man, Acta Unio. Int. Contra. Cancrum., 9:531-541 (1953).
61. D. H. Moore, D. H. Moore, and C. T. Moore. Breast-Carcinoma Etiological Factors, Advances in Cancer Research, 40:189-253 (1983).
62. S. P. Helmrich, S. Shapiro, L. Rosenberg, D. W. Kaufman, D. Slone, C. Bain, O. S. Miettinen, P. D. Stolley, N. B. Rosenshein, R. C. Knapp, T. Leavitt, D. Schottenfeld, R. L. Engle, and M. Levy. Risk-Factors for Breast-Cancer, American Journal of Epidemiology, 117:35-45 (1983).
63. J. L. Kelsey and G. S. Berkowitz. Breast-Cancer Epidemiology, Cancer Research, 48:5615-5623 (1988).
64. F. Salehi, L. Dunfield, K. P. Phillips, D. Krewski, and B. C. Vanderhyden. Risk factors for ovarian cancer: An overview with emphasis on hormonal factors, Journal of Toxicology and Environmental Health-Part B-Critical Reviews, 11:301-321 (2008).
65. J. L. Kelsey, M. D. Gammon, and E. M. John. Reproductive factors and breast cancer, Epidemiol. Rev., 15:36-47 (1993).
66. Spicer D.V. and Pike M.C. Epidemiology of breast cancer. 1994. New York, Raven Press. Ref Type: Generic
67. V. Cogliano, Y. Grosse, R. Baan, K. Straif, B. Secretan, and F. El Ghissassi. Carcinogenicity of combined oestrogen-progestagen contraceptives and menopausal treatment, Lancet Oncol., 6:552-553 (2005).
68. D. Chen, V. Milacic, M. Frezza, and Q. P. Dou. Metal complexes, their cellular targets and potential for cancer therapy, Curr. Pharm. Des, 15:777-791 (2009).
69. D. Chen, S. H. Song, M. G. Wientjes, T. K. Yeh, L. Zhao, M. Villalona- Calero, G. A. Otterson, R. Jensen, M. Grever, A. J. Murgo, and J. L. S. Au. Nontoxic suramin as a chemosensitizer in patients: Dosing nomogram development, Pharmaceutical Research, 23:1265-1274 (2006).
70. Y. Xin, G. Lyness, D. Chen, S. Song, M. G. Wientjes, and J. L. S. Au. Low dose suramin as a chemosensitizer of bladder cancer to mitomycin C, Journal of Urology, 174:322-327 (2005).
142 71. S. H. Song, B. Yu, Y. Wei, M. G. Wientjes, and J. L. S. An. Low-dose suramin enhanced paclitaxel activity in chemotherapy-Naive and paclitaxel-pretreated human breast xenograft tumors, Clinical Cancer Research, 10:6058-6065 (2004).
72. M. A. Fuertes, J. Castilla, C. Alonso, and J. M. Perez. Cisplatin biochemical mechanism of action: From cytotoxicity to induction of cell death through interconnections between apoptotic and necrotic pathways, Current Medicinal Chemistry, 10:257-266 (2003).
73. Walter Stumpf and Howard F.Solomon. Autoradiography and correlative imaging . 1995. San Diego , Academic Press. Autoradiography and correlative imaging . Walter Stumpf and Howard F.Solomon. Ref Type: Generic
74. E. C. Bradley, B. F. Issell, and R. Hellman. The Human-Tumor Colony- Forming Chemosensitivity Assay - A Biological and Clinical Review, Investigational New Drugs, 2:59-70 (1984).
75. R. Nunez. DNA measurement and cell cycle analysis by flow cytometry, Curr. Issues Mol. Biol., 3:67-70 (2001).
76. J. H. Walker, C. M. Boustead, J. J. Koster, M. Bewley, and D. A. Waller. Annexin V, a calcium-dependent phospholipid-binding protein, Biochem. Soc. Trans., 20:828-833 (1992).
77. K. A. Lewis, K. K. Lilly, E. A. Reynolds, W. P. Sullivan, S. H. Kaufmann, and W. A. Cliby. Ataxia telangiectasia and rad3-related kinase contributes to cell cycle arrest and survival after cisplatin but not oxaliplatin, Mol. Cancer Ther., 8:855-863 (2009).
78. D. N. Carney and C. F. Winkler. In vitro assays of chemotherapeutic sensitivity, Important Adv. Oncol.,78-103 (1985).
79. S. E. Salmon. Human tumor clonogenic assays: growth conditions and applications, Cancer Genet. Cytogenet., 19:21-28 (1986).
80. L. M. Weisenthal and M. E. Lippman. Clonogenic and nonclonogenic in vitro chemosensitivity assays, Cancer Treat. Rep., 69:615-632 (1985).
81. W. E. Stumpf. Drug localization and targeting with receptor microscopic autoradiography, J. Pharmacol. Toxicol. Methods, 51:25-40 (2005).
82. D. Zhang, M. Li, W. Ma, Y. Hou, Y. H. Li, S. W. Li, Q. Y. Sun, and W. H. Wang. Localization of mitotic arrest deficient 1 (MAD1) in mouse oocytes during the first meiosis and its functions as a spindle checkpoint protein, Biol. Reprod., 72:58-68 (2005).
143 83. C. C. Sprenger, R. H. Drivdahl, L. B. Woodke, D. Eyman, M. J. Reed, W. G. Carter, and S. R. Plymate. Senescence-induced alterations of laminin chain expression modulate tumorigenicity of prostate cancer cells, Neoplasia., 10:1350-1361 (2008).
84. S. H. Song, M. G. Wientjes, Y. Gan, and J. L. S. Au. Fibroblast growth factors: An epigenetic mechanism of broad spectrum resistance to anticancer drugs, Proceedings of the National Academy of Sciences of the United States of America, 97:8658-8663 (2000).
85. T. Boulikas and M. Vougiouka. Cisplatin and platinum drugs at the molecular level (review), Oncology Reports, 10:1663-1682 (2003).
86. M. S. Kim, S. L. Li, C. N. Bertolami, H. M. Cherrick, and N. H. Park. State of P53, Rb and Dcc Tumor-Suppressor Genes in Human Oral-Cancer Cell-Lines, Anticancer Research, 13:1405-1414 (1993).
87. L. J. Kuo and L. X. Yang. gamma-H2AX - A novel biomarker for DNA double-strand breaks, In Vivo, 22:305-309 (2008).
88. T. M. Mekhail and M. Markman. Paclitaxel in cancer therapy, Expert Opin. Pharmacother., 3:755-766 (2002).
89. H. E. Mischo, P. Hemmerich, F. Grosse, and S. S. Zhang. Actinomycin D induces histone gamma-H2AX foci and complex formation of gamma-H2AX with Ku70 and nuclear DNA helicase II, Journal of Biological Chemistry, 280:9586-9594 (2005).
90. Z. Lu, T. S. Wientjes, and J. L. Au. Nontoxic suramin treatments enhance docetaxel activity in chemotherapy-pretreated non-small cell lung xenograft tumors, Pharm. Res., 22:1069-1078 (2005).
91. Y. Gan, M. G. Wientjes, D. E. Schuller, and J. L. Au. Pharmacodynamics of taxol in human head and neck tumors, Cancer Res., 56:2086-2093 (1996).
92. S. H. MacPhail, J. P. Banath, T. Y. Yu, E. H. Chu, H. Lambur, and P. L. Olive. Expression of phosphorylated histone H2AX in cultured cell lines following exposure to X-rays, Int. J. Radiat. Biol., 79:351-358 (2003).
93. N. Taneja, M. Davis, J. S. Choy, M. A. Beckett, R. Singh, S. J. Kron, and R. R. Weichselbaum. Histone H2AX phosphorylation as a predictor of radiosensitivity and target for radiotherapy, J. Biol. Chem., 279:2273-2280 (2004).
94. S. H. MacPhail, J. P. Banath, Y. Yu, E. Chu, and P. L. Olive. Cell cycle- dependent expression of phosphorylated histone H2AX: reduced expression in unirradiated but not X-irradiated G1-phase cells, Radiat. Res., 159:759-767 (2003).
144 95. P. L. Olive, J. P. Banath, and L. T. Sinnott. Phosphorylated histone H2AX in spheroids, tumors, and tissues of mice exposed to etoposide and 3-amino-1,2,4- benzotriazine-1,3-dioxide, Cancer Res., 64:5363-5369 (2004).
96. T. Yu, S. H. MacPhail, J. P. Banath, D. Klokov, and P. L. Olive. Endogenous expression of phosphorylated histone H2AX in tumors in relation to DNA double-strand breaks and genomic instability, DNA Repair (Amst), 5:935-946 (2006).
97. E. L. Franco. Cancer Causes Revisited - Human Papillomavirus and Cervical Neoplasia, Journal of the National Cancer Institute, 87:779-780 (1995).
98. M. H. Schiffman, H. M. Bauer, R. N. Hoover, A. G. Glass, D. M. Cadell, B. B. Rush, D. R. Scott, M. E. Sherman, R. J. Kurman, S. Wacholder, C. K. Stanton, and M. M. Manos. Epidemiologic Evidence Showing That Human Papillomavirus Infection Causes Most Cervical Intraepithelial Neoplasia, Journal of the National Cancer Institute, 85:958-964 (1993).
99. H. H. zur. Molecular pathogenesis of cancer of the cervix and its causation by specific human papillomavirus types, Curr. Top. Microbiol. Immunol., 186:131-156 (1994).
100. C. K. Fairley, S. J. Chen, S. N. Tabrizi, M. A. Quinn, J. J. Mcneil, and S. M. Garland. Tampons - A Novel Patient-Administered Method for the Assessment of Genital Human Papillomavirus Infection, Journal of Infectious Diseases, 165:1103-1106 (1992).
101. M. H. Schiffman. Recent Progress in Defining the Epidemiology of Human Papillomavirus Infection and Cervical Neoplasia, Journal of the National Cancer Institute, 84:394-398 (1992).
102. J. M. M. Walboomers, M. V. Jacobs, M. M. Manos, F. X. Bosch, J. A. Kummer, K. V. Shah, P. J. F. Snijders, J. Peto, C. J. L. M. Meijer, and N. Munoz. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide, Journal of Pathology, 189:12-19 (1999).
103. M. Schiffman, R. Herrero, R. DeSalle, A. Hildesheim, S. Wacholder, A. C. Rodriguez, M. C. Bratti, M. E. Sherman, J. Morales, D. Guillen, M. Alfaro, M. Hutchinson, T. C. Wright, D. Solomon, Z. G. Chen, J. Schussler, P. E. Castle, and R. D. Burk. The carcinogenicity of human papillornavirus types reflects viral evolution, Virology, 337:76-84 (2005).
104. Chen W. HPV type-distribution in cervical cancer in China: a hospital- based multi-center study. Smith JS, Zhang X, Molijn A, Schmidt JE, Shi JF, Quint W, Jenkins D, Qiao YL, and Chinese HPV Typing Group. 8, PA2-05. 2007. Abstract Book, The 24th International Papillomavirus Conference and Clinical Workshop. Ref Type: Generic
145 105. K. A. Workowski and S. M. Berman. Sexually transmitted diseases treatment guidelines, 2006, MMWR Recomm. Rep., 55:1-94 (2006).
106. H. Wang and Y. L. Qiao. Human papillomavirus type-distribution in condylomata acuminata of mainland China: a meta-analysis, International Journal of Std & Aids, 19:680-684 (2008).
107. L. Gissmann, L. Wolnik, H. Ikenberg, U. Koldovsky, H. G. Schnurch, and H. H. zur. Human papillomavirus types 6 and 11 DNA sequences in genital and laryngeal papillomas and in some cervical cancers, Proc. Natl. Acad. Sci. U. S A, 80:560-563 (1983).
108. D. Dias, J. Van Doren, S. Schlottmann, S. Kelly, D. Puchalski, W. Ruiz, P. Boerckel, J. Kessler, J. M. Antonello, T. Green, M. Brown, J. Smith, N. Chirmule, E. Barr, K. U. Jansen, and M. T. Esser. Optimization and validation of a multiplexed Luminex assay to quantify antibodies to neutralizing epitopes on human papillomaviruses 6, 11, 16, and 18, Clinical and Diagnostic Laboratory Immunology, 12:959-969 (2005).
109. D. Opalka, C. E. Lachman, S. A. MacMullen, K. U. Jansen, J. F. Smith, N. Chirmule, and M. T. Esser. Simultaneous quantitation of antibodies to neutralizing epitopes on virus-like particles for human papillomavirus types 6, 11, 16, and 18 by a multiplexed luminex assay 2, Clinical and Diagnostic Laboratory Immunology, 10:108- 115 (2003).
110. J. J. Carter, L. A. Koutsky, G. C. Wipf, N. D. Christensen, S. K. Lee, J. Kuypers, N. Kiviat, and D. A. Galloway. The natural history of human papillomavirus type 16 capsid antibodies among a cohort of university women, J. Infect. Dis., 174:927- 936 (1996).
111. L. Wideroff, M. H. Schiffman, R. Hoover, R. E. Tarone, B. Nonnenmacher, N. Hubbert, R. Kirnbauer, C. E. Greer, A. T. Lorincz, M. M. Manos, A. G. Glass, D. R. Scott, M. E. Sherman, J. Buckland, D. Lowy, and J. Schiller. Epidemiologic determinants of seroreactivity to human papillomavirus (HPV) type 16 virus-like particles in cervical HPV-16 DNA-positive and-negative women, J. Infect. Dis., 174:937-943 (1996).
112. R. P. Viscidi, K. L. Kotloff, B. Clayman, K. Russ, S. Shapiro, and K. V. Shah. Prevalence of antibodies to human papillomavirus (HPV) type 16 virus-like particles in relation to cervical HPV infection among college women, Clin. Diagn. Lab Immunol., 4:122-126 (1997).
113. J. Dillner, I. Kallings, C. Brihmer, B. Sikstrom, P. Koskela, M. Lehtinen, J. T. Schiller, M. Sapp, and P. A. Mardh. Seropositivities to human papillomavirus types 16, 18, or 33 capsids and to Chlamydia trachomatis are markers of sexual behavior, J. Infect. Dis., 173:1394-1398 (1996).
146 114. A. O. Olsen, J. Dillner, K. Gjoen, and P. Magnus. Seropositivity against HPV 16 capsids: a better marker of past sexual behaviour than presence of HPV DNA, Genitourin. Med., 73:131-135 (1997).
115. B. Dondog, G. M. Clifford, S. Vaccarella, T. Waterboer, D. Unurjargal, D. Avirmed, S. Enkhtuya, F. Kommoss, N. Wentzensen, P. J. Snijders, C. J. Meijer, S. Franceschi, and M. Pawlita. Human papillomavirus infection in Ulaanbaatar, Mongolia: a population-based study, Cancer Epidemiol. Biomarkers Prev., 17:1731-1738 (2008).
116. K. M. Stone, K. L. Karem, M. R. Sternberg, G. M. McQuillan, A. D. Poon, E. R. Unger, and W. C. Reeves. Seroprevalence of human papillomavirus type 16 infection in the United States, J. Infect. Dis., 186:1396-1402 (2002).
117. Y. Y. Studentsov, G. Y. Ho, M. A. Marks, R. Bierman, and R. D. Burk. Polymer-based enzyme-linked immunosorbent assay using human papillomavirus type 16 (HPV16) virus-like particles detects HPV16 clade-specific serologic responses, J. Clin. Microbiol., 41:2827-2834 (2003).
118. A. Andersson-Ellstrom, J. Dillner, B. Hagmar, J. Schiller, M. Sapp, L. Forssman, and I. Milsom. Comparison of development of serum antibodies to HPV16 and HPV33 and acquisition of cervical HPV DNA among sexually experienced and virginal young girls. A longitudinal cohort study, Sex Transm. Dis., 23:234-238 (1996).
119. A. R. Kreimer, A. J. Alberg, R. Viscidi, and M. L. Gillison. Gender differences in sexual biomarkers and behaviors associated with human papillomavirus- 16, -18, and -33 seroprevalence, Sex Transm. Dis., 31:247-256 (2004).
120. C. J. Chen, R. P. Viscidi, C. H. Chuang, Y. C. Huang, C. H. Chiu, and T. Y. Lin. Seroprevalence of human papillomavirus types 16 and 18 in the general population in Taiwan: implication for optimal age of human papillomavirus vaccination, J. Clin. Virol., 38:126-130 (2007).
121. H. R. Shin, D. H. Lee, R. Herrero, J. S. Smith, S. Vaccarella, S. H. Hong, K. Y. Jung, H. H. Kim, U. D. Park, H. S. Cha, S. Park, A. Touze, N. Munoz, P. J. Snijders, C. J. Meijer, P. Coursaget, and S. Franceschi. Prevalence of human papillomavirus infection in women in Busan, South Korea, Int. J. Cancer, 103:413-421 (2003).
122. Y. P. Bao, N. Li, J. S. Smith, and Y. L. Qiao. Human papillomavirus type- distribution in the cervix of Chinese women: a meta-analysis, Int. J. STD AIDS, 19:106- 111 (2008).
123. Hu S. HPV16/18 high in precancer, low in normal cytology in China. Smith JS, Chen W, Ji J, Pretorius RG, Esser M, Taddeo FJ, Belinson JL, and Qiao YL. [30], 11. 2009. Abstract Book, The 25th International Papillomavirus Conference Clinical & Educational Workshop. Hu S., Smith JS, Chen W, Ji J, Pretorius RG, Esser M, Taddeo FJ, Belinson JL, and Qiao YL. 147 Ref Type: Generic
124. Markowitz L. Seroprevalence of HPV 6, 11, 16, 18 in United States. Sternberg M, Dunne EF, and Unger ER. 30, 2. 2009. Abstract Book, The 25th International Papillomavirus Conference Clinical & Educational Workshop. Markowitz L, Sternberg M, Dunne EF, and Unger ER. Ref Type: Generic
125. N. D. Christensen, R. Kirnbauer, J. T. Schiller, S. J. Ghim, R. Schlegel, A. B. Jenson, and J. W. Kreider. Human papillomavirus types 6 and 11 have antigenically distinct strongly immunogenic conformationally dependent neutralizing epitopes, Virology, 205:329-335 (1994).
126. J. Konya and J. Dillner. Immunity to oncogenic human papillomaviruses, Adv. Cancer Res., 82:205-238 (2001).
127. J. Dillner. The serological response to papillomaviruses, Semin. Cancer Biol., 9:423-430 (1999).
128. P. E. Castle, T. Shields, R. Kirnbauer, M. M. Manos, R. D. Burk, A. G. Glass, D. R. Scott, M. E. Sherman, and M. Schiffman. Sexual behavior, human papillomavirus type 16 (HPV 16) infection, and HPV 16 seropositivity, Sex Transm. Dis., 29:182-187 (2002).
129. S. Sukvirach, J. S. Smith, S. Tunsakul, N. Munoz, V. Kesararat, O. Opasatian, S. Chichareon, V. Kaenploy, R. Ashley, C. J. Meijer, P. J. Snijders, P. Coursaget, S. Franceschi, and R. Herrero. Population-based human papillomavirus prevalence in Lampang and Songkla, Thailand, J. Infect. Dis., 187:1246-1256 (2003).
130. S. S. Wang, M. Schiffman, T. S. Shields, R. Herrero, A. Hildesheim, M. C. Bratti, M. E. Sherman, A. C. Rodriguez, P. E. Castle, J. Morales, M. Alfaro, T. Wright, S. Chen, B. Clayman, R. D. Burk, and R. P. Viscidi. Seroprevalence of human papillomavirus-16, -18, -31, and -45 in a population-based cohort of 10000 women in Costa Rica, Br. J. Cancer, 89:1248-1254 (2003).
131. G. M. Clifford, H. R. Shin, J. K. Oh, T. Waterboer, Y. H. Ju, S. Vaccarella, W. Quint, M. Pawlita, and S. Franceschi. Serologic response to oncogenic human papillomavirus types in male and female university students in Busan, South Korea, Cancer Epidemiol. Biomarkers Prev., 16:1874-1879 (2007).
132. Nielsen A. High-risk HPV persistence in Danish women from the general population. Kjær S, Munk C, Osler M, and Iftner T. 30, 4. 2009. Abstract Book, The 25th International Papillomavirus Conference Clinical & Educational Workshop. Nielsen A, Kjær S, Munk C, Osler M, and Iftner T. Ref Type: Generic
148 133. J. L. Kelsey, M. D. Gammon, and E. M. John. Reproductive factors and breast cancer, Epidemiol. Rev., 15:36-47 (1993).
134. F. Salehi, L. Dunfield, K. P. Phillips, D. Krewski, and B. C. Vanderhyden. Risk factors for ovarian cancer: an overview with emphasis on hormonal factors, J. Toxicol. Environ. Health B Crit Rev., 11:301-321 (2008).
135. Ministry of Health. Year Book of Health in the People's Republic of China. 2004. Beijing, China, People's Health Press. Ref Type: Generic
136. National Family Planning Commission of P.R.China. Year Book of Population and Family Planning in the People's Republic of China. 2004. Beijing, China, China Population Press. Ref Type: Generic
137. National Family Planning Commission of P.R.China. Year Book of Family Planning in the People's Republic of China. 1990. Beijing, China, China Population Press. Ref Type: Generic
138. National Family Planning Commission of P.R.China. Year Book of Family Planning in the People's Republic of China. 2003. Beijing, China, China Population Press. Ref Type: Generic
139. Department of Population and Social Technology Statistics, National Bureau of Statistics. 97 China Population. 1998. Beijing, China, China Statistics Press. Ref Type: Generic
140. National Family Planning Commission of P.R.China. Year Book of Population and Family Planning in the People's Republic of China. 2008. Beijing, China, China Population Press. Ref Type: Generic
141. National Family Planning Commission of P.R.China. Year Book of Population and Family Planning in the People's Republic of China. 2005. Beijing, China, China Population Press. Ref Type: Generic
142. Statistical Information Center, Chinese Ministry of Health. Abstract of the report on the 3rd National Health Service Investigation and Analysis: Investigation and Research on China Health Service. 2004. Beijing, China, Peking Union Medical College Press. Ref Type: Generic
149 143. Liang JM. Statistics of 1988 National Fertility and Contraceptive sampling survey: the 3rd part: fertility and contraceptive. Chen SL. 1992. Beijing, China, China Population Press. Liang JM and Chen SL. Ref Type: Generic
144. Jiang ZH. Statistics of 1992 fertility sampling survey in China. 1996. Beijing, China, China Population Press. Ref Type: Generic
145. Jiang ZH. Data collection of 1997 national popultion and reproductive health survey. 2000. Beijing, China, China Population Press. Ref Type: Generic
146. Liang JM. Statistics of 1988 National Fertility and Contraceptive sampling survey: the 1st part: Population Structure. Chen SL. 1992. Beijing, China, China Population Press. Liang JM and Chen SL. Ref Type: Generic
147. Department of Population Statistics, National Bureau of Statistics. Statistics of 1987 one-per-hundredth population sampling survey in China: National Data. 1988. Beijing, China, China Statistics Press. Ref Type: Generic
148. Department of Population Statistics, National Bureau of Statistics. National Report of 1985 phase I intensive fertility survey in China: part I. 1986. Beijing, China, China Statistics Press. Ref Type: Generic
149. Department of Population Statistics, National Bureau of Statistics. Statistics of 1995 one-per-hundredth population sampling survey in China. 1996. Beijing, China, China Statistics Press. Ref Type: Generic
150. National Family Planning Commission of P.R.China. Statistics of 1988 National Fertility and Contraceptive sampling survey-National Data (Six Volumes): Volume I, Summed Data. 1989. Beijing, China, China Population Press. Ref Type: Generic
151. Pan GY. Statistics of 2001 National Family Planning and Reproductive Health sampling survey. 2004. Beijing, China, China Population Press. Ref Type: Generic
152. Research Institute of All China Women's Federation Research Office of Shanxi Provincial Women's Federation. Statistics on Chinese women, 1949-1989. 1991. Beijing, China, China Statistics Press. Ref Type: Generic
150 153. Yao XW. Fertility data of China. 1995. Beijing, China, China Population Press. Ref Type: Generic
154. C. Wu, R. M. Ray, M. G. Lin, D. L. Gao, N. K. Horner, Z. C. Nelson, J. W. Lampe, Y. W. Hu, J. Shannon, H. Stalsberg, W. Li, D. Fitzgibbons, P. Porter, R. E. Patterson, J. A. Satia, and D. B. Thomas. A case-control study of risk factors for fibrocystic breast conditions: Shanghai Nutrition and Breast Disease Study, China, 1995- 2000, Am. J. Epidemiol., 160:945-960 (2004).
155. Y. T. Gao, X. O. Shu, Q. Dai, J. D. Potter, L. A. Brinton, W. Wen, T. A. Sellers, L. H. Kushi, Z. Ruan, R. M. Bostick, F. Jin, and W. Zheng. Association of menstrual and reproductive factors with breast cancer risk: results from the Shanghai Breast Cancer Study, Int. J. Cancer, 87:295-300 (2000).
156. Pei GJ. Meta-analysis of risk factors of breast cancer in Chinese women. Fu L, Cui YL, Wang WL, Gao LH, and Lu WQ. Zhong Guo Fu You Bao Jian 23[19], 2650-2652. 2008. Pei GJ, Fu L, Cui YL, Wang WL, Gao LH, and Lu WQ. Ref Type: Generic
157. Zhang JB. Meta-analysis on risk factors of breast cancer. Feng XX. Zhong Guo Wei Sheng Tong Ji 19[3], 168-170. 2002. Zhang JB and Feng XX. Ref Type: Generic
158. Xiao YH. Meta-analysis of Chinese literatures on oral contraceptives and breast cancer. Yan ZH, Zhang WJ, and Han XW. Yao Wu Liu Xing Bing Xue Za Zhi 9[4], 179-181. 2000. Xiao YH, Yan ZH, Zhang WJ, and Han XW. Ref Type: Generic
159. Han YF. Meta-analysis of relationship between oral contraceptives and breast cancer. Wu B and Wang JQ. Qi Qi Ha Er Yi Yue Yuan Xue Bao 27[6], 703-704. 2006. Han YF, Wu B, and Wang JQ. Ref Type: Generic
160. S. Kummar, M. Gutierrez, J. H. Doroshow, and A. J. Murgo. Drug development in oncology: classical cytotoxics and molecularly targeted agents, Br. J. Clin. Pharmacol., 62:15-26 (2006).
161. Ji JJ. A validated assay for gamma-H2AX as a pharmacodynamic biomarker of response to DNA damage. Ravithat Putvatana, Yiping Zhang, Christophe Redon, Olga Sedelnikova, Sherry Yang, Yves Pommier, William Bonner, Robert Kinders, Ralph Parchment, Melinda Hollingshead, Jennifer Low, Anthony Murgo, Joseph Tomaszewski, and James Doroshow. Abstract #4027. 2007. Los Angeles, CA, 98th AACR Annual Meeting. Ji JJ, Ravithat Putvatana, Yiping Zhang, Christophe Redon, Olga Sedelnikova, Sherry Yang, Yves Pommier, William Bonner, Robert Kinders, Ralph
151 Parchment, Melinda Hollingshead, Jennifer Low, Anthony Murgo, Joseph Tomaszewski, and James Doroshow. Ref Type: Generic
162. S. Kummar, R. Kinders, L. Rubinstein, R. E. Parchment, A. J. Murgo, J. Collins, O. Pickeral, J. Low, S. M. Steinberg, M. Gutierrez, S. Yang, L. Helman, R. Wiltrout, J. E. Tomaszewski, and J. H. Doroshow. Compressing drug development timelines in oncology using phase '0' trials, Nat. Rev. Cancer, 7:131-139 (2007).
152