Medical Statistics for Cancer Studies

Trevor F. Cox is retired from Liverpool Cancer Trials Unit, University of Liverpool, UK

• 1 Introduction. 1.1. About Cancer. 1.2. Cancer studies. 1.3. R Code. 2. Cancer Biology and Genetics for Non-Biologists. 2.1. Cells. 2.2. DNA, Genes, RNA and Proteins. 2.3. Cancer – DNA Gone Wrong. 2.4. Cancer Treatments. 2.5. Measuring Cancer in the Patient. 3. Survival Analysis. 3.1. The Amazing Survival Equations. 3.2. Non-parametric Estimation of Survival Curves. 3.3. Fitting Parametric Survival Curves to Data. 3.4. Comparing Two Survival Distributions. 3.5. The ESPAC4-Trial. 3.6. Comparing Two Parametric Survival Curves. 4. Designing and Running a Clinical Trial. 4.1. Types of Trials and Studies. 4.2. Clinical Trials. 5. Regression Analysis for Survival Data. 5.1. A Weibull Parametric Regression Model. 5.2. Cox Proportional Hazards Model. 5.3. Accelerated Failure Time (AFT) Models. 5.4. Proportional Odds Models. 5.5. Parametric Survival Distributions for PH and AFT Models. 5.6. Flexible Parametric Models. 6. Clinical Trials: The Statistician’s Role. 6.1. Sample Size Calculation. 6.2. Examples of Sample Size Calculations; Phases I to III. 6.3. Group Sequential Designs. 6.4. More Statistical Tasks for Clinical Trials. 7. Cancer Epidemiology. 7.1. Measuring Cancer. 7.2. Cancer Statistics for Countries. 7.3. Cohort Studies. 7.4. Case-control Studies. 7.5. Cross-sectional Studies. 7.6. Spatial Epidemiology. 8. Meta-Analysis. 8.1. How to Carry Out a Systematic Review. 8.2. Fixed Effects Model. 8.3. Random Effects Model. 8.4. Bayesian Meta-analysis. 8.5. Network Meta-analysis. 8.6. Individual Patient Data. 9. Cancer Biomarkers. 9.1. Diagnostic Biomarkers. 9.2. Prognostic Biomarkers. 9.3. Predictive Biomarkers for Pancreatic Cancer. 9.4. Biomarker Trial Design. 10. Cancer Informatics. 10.1. Producing Genetic Data. 10.2. Analysis of Microarray Data. 10.3. Pre-processing NGS Data. 10.4. TCGA-KIRC: Renal Clear Cell Carcinoma.