Guogen Shan, Ph.D., is a professor in the department of biostatistics in the UF College of Public Health and Health Professions. He earned his master’s degree and Ph.D. in biostatistics from the State University of New York at Buffalo. Before coming to the University of Florida, he worked as a statistician at the Brain Trauma Foundation in New York and as an assistant and associate professor at the University of Nevada Las Vegas. He has received numerous awards and has published a book and numerous peer-reviewed papers. He has ongoing funding from the National Institute on Aging and National Cancer Institute (NCI).
His research interests are adaptive designs for cancer trials and Alzheimer’s disease trials, platform designs, meta-analysis, exact test, exact confidence interval, statistical methods for combat sports, empirical likelihood, ROC curve analysis, copula, nonparametric tests, permutation test, reliability test and efficient clinical designs.
What are your current research interests and/or what is a project you are currently working on?
I am interested in developing and applying new adaptive statistical designs to cancer trials to effectively select the subpopulations who benefit from the investigated drug(s) the most. Adaptive designs are flexible and effective to reduce sample sizes and costs, but they are often computationally intensive. I currently have a small NCI grant to develop novel adaptive randomized designs for cancer clinical trials with a binary endpoint built on my previous work on adaptive one-arm designs using numerical search algorithms. Recently, I started to work with other researchers at UF and other institutions to develop adaptive designs for cancer trials with a survival endpoint.
What do you want to achieve with your work and/or in your career?
We have developed several novel approaches to analyze data from cancer trials (e.g., bladder cancer, breast cancer, lung cancer) when the final sample size is not exactly the same as the planned value. In such scenarios, the prespecified threshold values to decide futility or efficacy cannot be applied directly as the sample sizes have changed. Proper and effective statistical methods should be developed to analyze data from the trials with sample size change or adaptive designs. With the possibility of more patients in the investigated drug group when the drug is effective as compared to the gold standard, new statistical methods should be developed and applied for data analysis. We are working on our long-term goal of developing adaptive designs and proper statistical methods for data analysis. Meanwhile, it is equally important to develop software programs to share with other researchers.
What excites you about your work? What is exciting to you about your field right now?
I am extremely excited to find out that the methods we developed are used in real cancer trials for study design and/or data analysis. With assistance from supercomputing (e.g., HiPerGator at UF), more adaptive designs are being developed to meet the challenges from cancer trials. It is exciting to see more and more adaptive designs being used in cancer trials, such as response adaptive randomization, adaptive enrichment designs and promising zone designs.
“It is exciting to see more and more adaptive designs being used in cancer trials, such as response adaptive randomization, adaptive enrichment designs and promising zone designs.”
Guogen Shan, Ph.D.
What do you like to do outside of work?
My hobbies include tennis and hiking. It is so much fun to play tennis/sports with family and friends. We really enjoy the weather in Florida with a lot of outdoor activities. After we moved to Florida, we went to beaches frequently during the summer.