Statistics Curriculum Steering Committee:
Jordan Kharofa, MD – University of Cincinnati
Kara Leonard, MD, MS – Brown University
Bailey Nelson, MD – University of Cincinnati
Session 1 – Basic Biostatistics Concepts for Radiation Oncologists
Kara Leonard, MD
Assistant Professor, Department of Radiation Oncology
Warren Alpert Medical School of Brown University
Supplemental material for Session 1
Survival Analysis Excel Sheet
Survival Analysis Worksheet
0:00 -Introduction
2:25- Sensitivity and Specificity, Positive/Negative Predictive Value
21:10- Positive and Negative Likelihood Ratios
28:00- Receiver/Operator Curves
29:33- Pre- and Post-Test Probability
39:35- Survival Analysis
41:50- Kaplan-Meier Curves
1:00:19- Log-Rank Test
1:05:14- Wilcoxon Rank Sum
1:05:32- Cox Proportional Hazard Model
Session 2 – Thinking Better About Cancer Drugs
Vinay Prasad, MD, MPH
Associate Professor, Department of Medicine
University of California, San Francisco
Session 3 – Basic Intro Biostatistics for Radiation Oncology Residents
Clifton D. Fuller, MD, PhD
Associate Professor, Department of Radiation Oncology
MD Anderson Cancer Center
Supplemental material for Session 3
0:00- Introduction
5:10-Types of Measurement Scales
6:24- Summarizing and Displaying Data
9:29- Probability Distributions
14:07- Measures of Central Tendency, Measures of Spread
19:00-Graphical Representation
27:44- Sampling Variability, Standard Error
29:00- Confidence Intervals
29:53- Sample Size and Variability of Mean
30:45- Significance Testing
48:45- Statistical Error
51:00- t-Tests
53:45- Means of More than Two Samples
55:50- Multiple Comparisons and p-values
56:30- Mann-Whitney Test
58:05- Kruskal-Wallis H-Test
Session 4 – Non-experimental Studies and Observational Data
David Sher, MD, MPH
Associate Professor, Department of Radiation Oncology
UT Southwestern
0:00- Introduction
6:12- Bias
8:15- Confounding
11:18- Effect Modification
12:09- Fundamental Comparisons
14:38- Univariable Calculations
15:54- Fisher Exact Test
20:01- Chi-squared Test
26:32 Pearson Correlation
28:30- Spearman Correlations
30:33- Linear Regression
34:58- Odds Ratio, Risk Ratio
36:39- Logistic Regression
40:58- Building a Model
45:59- Propensity Score Analysis
47:59- Instrumental Analysis
50:50- Case-Control Studies
53:10- Cross-Sectional Study
55:11- Cohort Study
Session 5 – Radiobiologic Principles in RO Clinical Research Design
Søren Bentzen, PhD, DMSC
Director of the Division of Biostatistics and Bioinformatics, Department of Epidemiology and Public Health
University of Maryland School of Medicine
0:00 Introduction
2:21 Discussion of historic RO trial (Copenhagen Breast Cancer Trial)
4:06 Summary of UK post-op RT trials for early breast cancer
6:08 Classical hypothesis testing
12:04 Phases of clinical trials
17:19 Cochrane ladder of evidence
19:19 What’s special about RO trials?
19:19 Double Blinding
19:19 Phase I/II or Feasibility Trials
19:19 Drug-Radiation Combinations
29:28 Toxicity monitoring
32:29 Dose-response relationship and trial design
43:44 Choice of endpoints
Session 6 – Bayesian Statistics in Clinical Trials
Peter Thall, MD
Professor, Department of Biostatistics
MD Anderson Cancer Center
0:00 Introduction
2:20 Elements of Bayesian Statistics
4:35 Beta Distributions
5:40 Simple Example
10:17 Estimating Probabilities of Rare Events
11:19 Simple Example, continued
14:38 Example: Kaplan Meier Estimates of PFS Time Distribution for Each Dose from a Phase I-II Trial of Targeted Agent in Advanced Renal Cancer
18:15 Quantifying Strength of Evidence
28:28 p-values
31:50 Looking at Efficacy and Toxicity- A Simple Example
34:49 Utility Based Clinical Trial Design
37:47 Example: A Radiation Dose-Finding Trial in Pediatric Brain Tumors
42:42 Example: Randomized Subgroup-Specific Comparison of NuPrehab vs Standard of Care Effects on Post-op Morbidity after Esophageal Resection
47:47 Example: PK-Guided vs. Fixed IV Busulfan Dose in Allogeneic Stem Cell Transplantation
52:26 Example: Randomized Pilot Study of T-reg Cell Therapy for ARDS in Intubated COVID-19 Patients
Session 7 – Machine Learning Models for Radiation Oncology
John Kang, MD, PhD
Department of Radiation Oncology
University of Washington
0:00 Introduction
1:48 Machine learning and statistics
3:35 Modeling Nature
8:20 Examples of real-life models
12:41 Current state of AI in Rad Onc
14:51 Deep learning for diagnosis and treatment guidance
17:07 Definitions
20:11 Importance of using the correct model
24:37 Clinical models
26:38 Tips for creating a machine learning nomogram
27:57 Random Forest Model
33:15 Improving on Kattan-type models
36:07 Case study
47:22 ROC curves
Session 8 – Analyzing and Interpreting Patient Reported Outcomes
Benjamin Movsas, MD
Professor and Chair, Department of Radiation Oncology
Henry Ford Cancer Institute
0:00 Introduction
2:13 Definition of PRO
4:08 “Overcoming the Cons”
6:23 Secondary Analysis of RTOG 0617
21:30 RTOG 9801 QOL analysis
22:43 “Disadvantage of Men Living Alone Participating in RTOG Head and Neck Trials”
24:10 Quality of Life
28:25 PROs and PCI (RTOG 0214)
29:55 Suggested References
32:38 Lung SBRT and PROs
34:22 QOL Instruments
36:50 PRO-CTCAE
38:22 Collecting QOL data
41:15 Future of PROs/QOL
45:30 PROceeding with PROs
