Cancer Mortality and Morbidity Patterns in the U.S. Population: An Interdisciplinary Approach (Statistics for Biology and Health)

This book is the first of its kind to describe interdisciplinary approaches to biomedical studies. It views analyses of biomedical data sets, such as cancer morbidity and mortality, from a different and richer than classic epidemiological perspective by using mathematical modeling methods, including ones providing insights into probable mechanisms of human carcinogenesis. The book will be useful for many specialists, e.g., epidemiologists, oncologists, medical researchers, biologists, public health and environmental specialists, and specialists in mathematical modeling. Medical, biology and math undergraduates and postgraduates, as well as basic and applied researchers attempting to extend their studies in collaboration with other specialists in interdisciplinary teams, will find practical information here. Biomedical specialists could be interested in historical aspects of cancer treatment and prevention, mechanisms of carcinogenesis, cancer risk factors, cancer mortality and morbidity trends in the U.S. over a more than 50-year period, as well as specific features of cancer histotypes, and recent approaches to cancer prevention. Readers interested in analytic aspects can find information on existing and innovative approaches used in interdisciplinary studies such as stochastic process models, microsimulation of interventions, and empirical Bayes approaches. This book was written by authors with different backgrounds who teamed in an interdisciplinary group. Kenneth G. Manton, Ph. D. (Demography) is Research Professor of Demographic Studies at Duke University (Durham, NC). He was Head of the W.H.O. Collaborating Center for Research and Training in the Methods of Assessing Risk and Forecasting Health Status Trends. He has authored more than 450 peer-reviewed publications, including several books. Igor Akushevich, Ph. D. (Theoretical and Mathematical Physics), Center for Population Health and Aging at Duke University, authored more than 70 peer-reviewed publications. Julia Kravchenko, MD, Ph. D. (Internal Diseases, Biochemistry), Duke Comprehensive Cancer Center in the School of Medicine. She is author of more than 30 peer-reviewed publications. Cover 1 Frontmatter 2 Statistics for Biology and Health 2 Foreword 7 Preface 8 Contents 10 Cancer Contra Human: Cohabitation with Casualties? 15 1.1 Endless Assault: Prehistory and History of Human Cancer 15 1.2 Global and U.S. Cancer Morbidity and Mortality Trends: Historic Perspectives 20 1.2.1 Global Cancer Morbidity and Mortality: At the Beginning of 21st Century 20 1.2.2 U.S. Cancer Morbidity and Mortality: At the Beginning of 21st Century 26 1.2.3 Cancer Mortality: U.S. Historical Trends 35 1.3 Interdisciplinary Approach to Population Health Studies 45 References 48 Cancer Modeling: How Far Can We Go? 51 2.1 Cellular Aspects of Carcinogenesis 52 2.1.1 Nuclear DNA Mutation 52 2.1.2 Mitochondrial DNA Mutation 53 2.1.3 Damage to the Protein Generation Machinery of the Endoplasmic Reticulum and Golgi Apparatus 55 2.1.4 Cell-Cell Communication 56 2.1.5 Telomere Control of Cell Division 57 2.1.6 Apoptosis 58 2.1.7 Angiogenesis 59 2.1.8 Immunomodulation 59 2.1.9 Metalloproteinases 60 2.2 Theories of Carcinogenesis 60 2.2.1 Somatic Mutation Theory of Carcinogenesis 62 2.2.2 The Stem Cell Theory of Carcinogenesis 62 2.2.3 Mutation versus Epigenetic Theories of Carcinogenesis 62 2.2.4 The Tissue Organization Field Theory of Carcinogenesis and Neoplasia 63 2.2.5 Telomere Dysfunction Theory 63 2.3 An Overview of Formal Quantitative Models of Carcinogenesis 63 2.3.1 Nordling, and Armitage and Doll 67 2.3.2 The Moolgavkar-Venzon-Knudson Two-Stage Model 69 2.3.3 The Generalized MVK and Armitage-Doll Models 72 2.3.4 The Multiple Pathway Models of Carcinogenesis 73 2.3.5 Mixed Models of Carcinogenesis 74 2.3.6 Cancer at Old Age and Approaches to Modeling: If the Cancer Incidence Rates Are Declining? 75 2.3.7 Complexity and Chaos Theory 78 2.3.8 Statistical/Empirical Cancer Models 78 2.3.9 The Other Common Modeling Approaches for Carcinogenesis 79 2.4 Modeling for Populations with Heterogeneous Mutational Events and Tumor Growth Rates 80 2.4.1 Model Innovations, Fuzzy State Processes, Heterogeneous Tumor Risks, and Tumor Growth Rates 80 2.4.2 Intracellular Processes: Interactions Complicate Modeling 80 2.4.3 Tumor Growth and Growth Heterogeneity 85 2.4.4 Stochastic Multivariate Models of Carcinogenesis 87 2.5 Summary 88 References 90 Cancer Risk Factors 102 3.1 Overview of Cancer Risk Factors 102 3.1.1 Biomarkers 102 3.1.2 Genotoxic and Nongenotoxic Mechanisms 103 3.1.2.1 Mutations and DNA Damage: Exogenous and Endogenous Agents 105 3.1.2.2 What Modulates the Rate of Carcinogenesis in Individuals? 106 3.1.2.3 How Nongenotoxic Carcinogens Work 107 3.1.2 Controllable and Noncontrollable Risk Factors 107 3.1.3.1 Results of Meta-analyses for Controllable Risk Factors 115 3.1.3.2 Limitations of Meta-analyses 115 3.1.3.3 Surrounded by Cancer Risk Factors: Rumors or Facts? 126 3.1.3 Advanced Age as Cancer Risk Factor 128 3.1.3 Factors to Consider in Cancer Risk Analysis and Cancer Risk Prediction 129 3.2 Environmental Cancer Risk Factors 131 3.2.1 Radiation Exposure 132 3.2.1.1 Ionizing Radiation and Cancer Risk 136 3.2.1.2 Ionizing Radiation and Thyroid Cancer 140 Age-Dependent Prevalence 144 Latent Period 145 Age-Sex Differences in Thyroid Cancer Risk 145 3.2.2 Nutrition as a Cancer Risk Factor 147 3.2.2.1 Cooking Process 151 3.2.2.2 Acrylamide 151 3.2.2.3 Artificial Sweeteners 152 3.3 Summary 153 References 153 Standard and Innovative Statistical Methods for Empirically Analyzing Cancer Morbidity and Mortality 163 4.1 Survival Analysis and Life Table Models 164 4.1.1 Useful Modification of Standard Life Table Computations 167 4.1.2 Standard Regression Procedures and Some of Their Limitations in Describing Longitudinal Data 170 4.2 Multiple Disease Stochastic Compartment Models for Complex Cancer Population Mortality Curves: A Two-Disease Analysis of U.S. Female Breast Cancer 173 4.3 Stochastic Process Models of Cancer Risk: Latent and Observed State Variables Dependence 176 4.3.1 Microsimulation Estimation of Stochastic Process Parameters 182 4.4 Evaluation of Characteristics of Individual and Grouped Data 187 4.4.1 Detection of Disease Onset 189 4.4.2 Cancer Incidence in U.S. Elderly 189 4.4.3 Sensitivity Analysis 191 4.4.4 Uncertainty in Onset Calculation 191 4.4.5 Medicare Coverage and Censoring Uncertainties 191 4.4.6 Age Reporting Uncertainties 192 4.5 Generalized Frailty Model 194 4.6 Summary 197 References 198 Stochastic Methods of Analysis 202 5.1 Introduction 202 5.2 Stochastic Process Models 204 5.3 Microsimulation and Interventions 207 5.4 Grade of Membership and Other Latent-State Analysis Methods 210 5.4.1 GoM Model 213 5.4.2 LLS Model 214 5.5 A Geo-epidemiological/Mapping Study of the U.S. Cancer Mortality Rates and Trends Based on an Empirical Bayes Approach 216 5.6 Summary 223 References 224 U.S. Cancer Morbidity and Mortality Trends 228 6.1 Introduction: Cancer Mortality and Morbidity Registration 228 6.2 U.S. Cancer Mortality 229 6.3 U.S. Cancer Incidence 234 6.4 Morbidity and Mortality of Specific Cancer Sites 243 6.4.1 Lung Cancer 243 6.4.2 Breast Cancer 246 6.4.3 Prostate Cancer 246 6.4.4 Colorectal Cancer 247 6.4.5 Esophageal Cancer 248 6.4.6 Stomach Cancer 248 6.4.7 Cancer of Pancreas 249 6.4.8 Cancer of Liver and Intrahepatic Duct 249 6.4.9 Cancer of Corpus Uteri 250 6.4.10 Cervical Cancer 251 6.4.11 Kidney Cancer 251 6.4.12 Thyroid Cancer 252 6.5 Summary 252 References 253 U.S. Cancer Morbidity: Modeling Age-Patterns of Cancer Histotypes 258 7.1 Introduction 258 7.2 Analyses of Trends of Cancer Histotypes in the U.S. Population 259 7.2.1 Lung Cancer 259 7.2.2 Esophageal Cancer 267 7.2.3 Stomach Cancer 268 7.2.4 Colorectal Cancer 268 7.2.5 Cancer of Pancreas 269 7.2.6 Liver Cancer 269 7.2.7 Breast Cancer 270 7.2.8 Cancer of Uterus 270 7.2.9 Ovarian Cancer 271 7.2.10 Cancer of Cervix Uteri 271 7.2.11 Prostate Cancer 272 7.2.12 Cancer of Kidney 272 7.3 Analyses of Selected Cancer Histotypes for the Two-Disease Model 273 7.3.1 Breast Cancer: If Genetic Background May Result in Two Forms of Disease? 290 7.3.2 Cervical Cancer: Age Periods of Increased Susceptibility and Cancer Risk 294 7.3.3 Hepatocellular Carcinoma: Behavioral Risks and Viral Hepatitis Infections 301 7.3.4 Prostate Cancer: Screening Effects, Genetic Predisposition, or Something Else? 304 7.4 Squamous Cell Carcinomas and Adenocarcinomas: The Time Trends 306 7.5 Summary 309 References 311 Risk Factors Intervention 323 8.1 Environmental Risk Factor Contribution to Cancer and Noncancer Diseases 323 8.1.1 Health-Based Population Forecasting Effects of Smoking on Mortality from Cancer and Noncancer Diseases 325 8.1.1.1 Life Expectancy 332 8.1.1.2 Cancer and Noncancer Disease Contributions to Mortality 332 8.1.1.3 Sensitivity Analysis 333 Base Year and Time Trend of Mortality and Birth Rates 335 Time Trends of Smoking Rates 335 Uncertainties of Relative Risks 336 Former Smokers 336 8.1.1.4 Summary 336 8.2 Atherosclerosis and Cancer: Anything in Common? 337 8.3 Making Projections of Cardiovascular Disease Risk 343 8.4 Discussion and Comparison of the Results of the Two Intervention Studies 348 8.5 Summary 351 References 353 Cancer Prevention 361 9.1 Brief Overview of Prevention History, Strategies, Conquests, and Uncertainty 361 9.2 Smoking 369 9.3 Diet 372 9.4 Obesity 381 9.5 Physical Activity 382 9.6 Alcohol 383 9.7 Nonsteroidal Antiinflammatory Drugs 383 9.8 Statistical and Modeling Approaches for Prognoses 383 9.9 The Present and Future of Microsimulation Models in Tasks of Cancer Prevention 390 9.10 Summary 391 References 393 Backmatter 403 Conclusion and Outlook 403 Keywords 405 Abbreviations 406 Glossary 411 Index 452

This book is the first of its kind to describe interdisciplinary approaches to biomedical studies. It views analyses of biomedical data sets, such as cancer morbidity and mortality, from a different and richer than classic epidemiological perspective by using mathematical modeling methods, including ones providing insights into probable mechanisms of human carcinogenesis.

The book will be useful for many specialists, e.g., epidemiologists, oncologists, medical researchers, biologists, public health and environmental specialists, and specialists in mathematical modeling. Medical, biology and math undergraduates and postgraduates, as well as basic and applied researchers attempting to extend their studies in collaboration with other specialists in interdisciplinary teams, will find practical information here. Biomedical specialists could be interested in historical aspects of cancer treatment and prevention, mechanisms of carcinogenesis, cancer risk factors, cancer mortality and morbidity trends in the U.S. over a more than 50-year period, as well as specific features of cancer histotypes, and recent approaches to cancer prevention. Readers interested in analytic aspects can find information on existing and innovative approaches used in interdisciplinary studies such as stochastic process models, microsimulation of interventions, and empirical Bayes approaches.

This book was written by authors with different backgrounds who teamed in an interdisciplinary group. Kenneth G. Manton, Ph.D. (Demography) is Research Professor of Demographic Studies at Duke University (Durham, NC). He was Head of the W.H.O. Collaborating Center for Research and Training in the Methods of Assessing Risk and Forecasting Health Status Trends. He has authored more than 450 peer-reviewed publications, including several books.

Igor Akushevich, Ph.D. (Theoretical and Mathematical Physics), Center for Population Health and Aging at Duke University, authored more than 70 peer-reviewed publications.

Julia Kravchenko, MD, Ph.D. (Internal Diseases, Biochemistry), Duke Comprehensive Cancer Center in the School of Medicine. She is author of more than 30 peer-reviewed publications.

Front Matter....Pages i-xv Cancer Contra Human: Cohabitation with Casualties?....Pages 1-36 Cancer Modeling: How Far Can We Go?....Pages 37-87 Cancer Risk Factors....Pages 89-149 Standard and Innovative Statistical Methods for Empirically Analyzing Cancer Morbidity and Mortality....Pages 151-189 Stochastic Methods of Analysis....Pages 191-216 U.S. Cancer Morbidity and Mortality Trends....Pages 217-246 U.S. Cancer Morbidity: Modeling Age-Patterns of Cancer Histotypes....Pages 247-311 Risk Factors Intervention....Pages 313-350 Cancer Prevention....Pages 351-392 Back Matter....Pages 393-455 The purpose of this book is to examine the etiology of cancer in large human populations using mathematical models developed from an inter-disciplinary perspective of the population epidemiological, biodemographic, genetic and physiological basis of the mechanisms of cancer initiation and progression. In addition an investigation of how the basic mechanism of tumor initiation relates to general processes of senescence and to other major chronic diseases (e.g., heart disease and stroke) will be conducted.