2022 Past Events
2Q Event: Professional Seminar (June 16, 12-1pm PST)
Speaker: Dr. Gene Pennello
Senior Mathematical Statistician, Division of Imaging Diagnostics and Software Reliability, Food and Drug Administration
Abstract:
We show how to set classification accuracy goals for diagnostic tests based on the test meeting acceptable levels of risk stratification, i.e., the risk of having the disease condition post-test vs. pre-test (predictive value vs. prevalence). For a desired risk stratification, we use Bayes Theorem to determine goals for negative or positive likelihood ratios (NLR, PLR). Goals for specificity (Spec) and sensitivity (Sens) are implied by the NLR and PLR goals, but are less flexible. Because NLR, PLR, Spec, and Sens do not depend on prevalence, goals for them are especially useful for validating tests of low prevalence conditions in moderately-sized case-control studies enriched with subjects having the condition. For comparative diagnostic studies, goals that confer that a test is superior or non-inferior to a comparator are based on approximating risk differences and relative risks by functions of likelihood ratios. Wald confidence intervals are developed to facilitate hypothesis testing. To illustrate, we consider a fetal fibronectin assay for ruling out risk of pre-term birth and two human papillomavirus assays for ruling in cervical cancer. We also provide some examples of how to extract acceptable levels of risk stratification from various guidelines on medical decision making.
2Q Event: Chapter Picnic at Pacific Highlands Ranch Community Park (May 21, 2022 4:30pm -7:00pm)
1Q Event: SDASA 2022 Professional Seminar (Virtual): Biostatistics in Genomic Life Sciences Presented by Illumina, Inc.
When: Thursday, March 3, 2022, 4 PM to 5 PM Where: Virtual, Microsoft Teams
1. “About Illumina”
Kristen Meier
Senior Director, Biostatistics
2.“Diagnostics in the Era of Precision Medicine”
Johan Surtihadi
Director, Biostatistics
3. “Diagnostics Test Performance”
Darcy Vavrek
Senior Staff Biostatistician, Biostatistics
2021 Past Events
4Q Event:
|
ASA San Diego Chapter Seminar
Adaptive 2-in-1 design, expansion decision and extensions
|
|
Date/Time
|
Friday, Oct 22, 2021 from 2:00 PM- 3:00 PM PST
|
|
Abstract
|
The 2-in-1 adaptive design [Chen et al. 2018] allows seamless expansion of an ongoing Phase 2 trial into a Phase 3 trial to expedite a drug development program. Under a mild assumption that is expected to generally hold in practice, the trial can be tested at the full alpha level with or without expansion. The endpoint used for expansion decision can be the same as or different from the primary endpoints, and there is no restriction on the expansion bar. Due to its flexibility and robustness, it has drawn immediate attention to academic researchers and industry practitioners. The design has been substantially extended in the literature and successfully implemented in multiple trials.
In this presentation, we will discuss about decision on the expansion bar as well as various extensions of the 2-in-1 design.
|
|
Instructors
|
Dr. Cong Chen, Executive Director at Merck & Co., Inc.
|
SDASA 2021 Picnic
The picnic was hosted at the pacific highlands ranch community park on September 19 from 4:30 to 6:30 pm. About 40 people showed up and the event was well-received. We used the catering from Panera bread. The event was well received and people were happy to get to see each other after the Covid pandemic.
SDASA 2021 Professional Seminar and Chapter Business Meeting [slides] (Virtual)
Seminar: A Comprehensive and Practical Tutorial on Dose-Finding Trial Designs, mTPI, CRM, BOIN, or i3+3?
Presenter: Dr. Yuan Ji
When: Monday, May 24, 2021, 2:30 PM to 5 PM
Where: Virtual
Abstract
In the last three decades, numerous innovative and powerful dose-finding designs have been developed. More recently, many review papers are published claiming different designs performing superior than others. Software tools have also been made available, greatly facilitating the practical applications of these designs. In this talk, I will provide theoretical and practical review of some representative designs, including the interval-based designs such as mTPI, i3+3, and BOIN, and model-based designs like CRM. Interestingly, a common decision framework can unify most of the popular designs, which will be introduced. I will also touch upon real-world applications and some feedbacks from interactions with regulatory agents. A commercial tool will be introduced and demonstrated.
Speaker's Bio:
Dr. Yuan Ji is Professor of Biostatistics at The University of Chicago. He is an NIH-funded PI focusing on innovative computational and statistical methods for translational cancer research. Dr. Ji is author of over 140 publications in peer-reviewed journals, conference papers, book chapters, and abstracts, including Nature, Nature Methods, JCO, JNCI, JASA, and Biometrics, across medical and statistical journals. He is the inventor of many innovative Bayesian adaptive designs such as the mTPI and i3+3 designs, which have been widely applied in dose-finding clinical trials worldwide, including trials published on Lancet Oncology, JAMA oncology and JCO. His work on cancer genomics has been reported by a large number of media outlets in 2015. In particular, he led a publication in Nature Methods and invention of a tool called TCGA-Assembler which has been downloaded over 10,000 times worldwide. His recent work on precision medicine was elected as one of the top 10 ideas of the Precision Trials Challenge hosted by The Harvard Business School in 2015. He received Mitchell Prize in 2015 by the International Society for Bayesian Analysis. He is an elected fellow of the American Statistical Association.
2020 Past Events
SDASA 2020 4Q Event: Short Course on Oct 29 1-4pm PDT (online)
SDASA Sponsored Short Course: Robust Design and Analysis of Clinical Trials with Non-proportional Hazards: Methodology and Implementation in R
When: Oct 29 1-4pm PDT (online short course)
Fee: Free for ASA San Diego Chapter(SDASA) members only. For non San Diego chapter members, please sign up for 1-year membership ($10 for regular member and $5 for student) before registering (follow this link to become an ASA San Diego Chapter member).
Link to short course registration by Oct 28: Eventbrite Registration Link
You must register in order to receive the webex link.
About this short course:
Instructors: Satrajit Roychoudhury (Pfizer Inc) and Keaven Anderson (Merck & Co., Inc)
Overview:
Targeting the immune system has emerged as a promising treatment for cancer patients in recent years. Immunotherapy boosts the body's natural defenses to fight cancer. However, this novel treatment poses new challenges in the study design and statistical analysis of clinical trials. A major challenge is the delayed onset of treatment effects due to the mechanism of immunotherapy which violates the proportional hazard (PH) assumption. The conventional log-rank test may suffer a significant power loss in such scenarios.
This course will focus on different alternative design and analysis approaches for immune-oncology trials with potential non-proportional data. These include piecewise log-rank test and average hazard ratio, weighted log-rank test, combination tests and Kaplan-Meier based methods. The short course will also introduce a new approach based on a combination of multiple Fleming-Harrington WLR tests and is referred as the MaxCombo test which adaptively chooses best test depending on the underlying data. Finally, a stepwise approach for estimating treatment effect.
The short course will provide the analysis methodology, general design strategies, sample size calculation, evaluation of operating characteristics and necessary steps for protocol implementation in real life. The course with use real life examples and practical implementation with the new R package simtrial, gsDesign2 and gsdmvn.
PRESENTER(S):
Dr. Satrajit Roychoudhury is a Senior Director and a member of Statistical Research and Innovation group in Pfizer Inc. Prior to joining, he was a member of Statistical Methodology and Consulting group at Novartis. He started his career as a research statistician in Schering-Plough Research Institute (now Merck Co.). He has 12+ years of extensive experience in working in different phases of clinical trials. His primary expertise includes implementation of innovative statistical methodology in clinical trials. He has co-authored several publications/book chapters in this area and provided statistical training at major conferences. His areas of research include the use of survival analysis, model-informed drug development and Bayesian methods in clinical trials.
Dr. Keaven M Anderson is a Associate Scientific VP heading the Methodology Research biostatistics group at Merck where he has worked since 2003. He is a fellow of the ASA. Since completing his PhD in mathematical statistics at Stanford University and a post doc at the Harvard School of Public Health, Keaven has also worked at the Framingham Heart Study for the NIH prior to taking a pharmaceutical position at Centocor/J&J. He has long-standing interests in group sequential design and survival analysis. More recently, he has been interested in applications of multiplicity control using graphical methods and in trial design for immuno-oncology trials.
2020 Annual Business Meeting (Slides) and professional seminar
Guest Speaker: Dr. Linda Sun from Merck (Slides), presentation title: Two Innovative Designs for Confirmatory Oncology Trials
January 28, 2020, at Pfizer La Jolla
In this talk, we will discuss two innovation designs for confirmatory oncology trials in answering the call of efficiency in trial design given the fast advancement of science. One is to test monotherapy and combination therapy in one Phase III trial with biomarker consideration (Sun 2019). It is a common scenario that an experimental oncology therapy, as a monotherapy, may be more effective than standard of care (SOC) in a biomarker positive population but less so or even inferior to SOC in biomarker negative population. At the same time, due to synergistic or additive effect, the combination of the two may be more effective than SOC alone in the all-comer population. The conventional development paradigm is to conduct two separate Phase III trials, one with the monotherapy versus SOC in the biomarker positive population, and the other with the combination therapy versus SOC in the all-comer population. We propose a one-trial design that stratifies by biomarker status and randomizes biomarker positive patients into three arms (combination therapy, monotherapy, and SOC) and biomarker negative patients into two arms (combination therapy and SOC). The second design which we will discuss is a seamless Phase II/III design called two-in-one design (Chen 2018). We will introduce what two-in-one design is and how to use intermediate endpoints to make cost effective Go-No-Go decision in this seamless Phase II/III design.
2019 Past Events
2019 ASA Traveling Course: Big Data, Data Science and Deep Learning for Statistician
Oct 19, 2019, at SDSU
2017 Past Events
2017 Annual Business Meeting
June 27, 2017, at Pfizer La Jolla.
2016 Past Events
2016 Business Annual Meeting
Guest Speaker: Dr. Mingan Yang from San Diego State University (SDSU)
April 27, 2016, Pfizer La Jolla
ASA Traveling Course 2015
Year in review was given by outgoing president Cleridy Lennert, the financial status was presented by acting treasurer Patricia English, and the perspective for 2017 was provided y incoming president Patricia English.
Instructor: Dr. Dianne Cook, Professor of Statistics, Iowa State University dicook@iastate.edu
Course: Working with R to Analyze and Plot Data (hands on course).
Time: 8:30AM to 5PM May 18, 2015 (Monday) (Class starts at 9AM), Pfizer La Jolla Campus
This workshop focuses on a data centric introduction to using R, in a reproducible way, incorporating lots of data graphics and exploratory data analysis. The sections will center around contemporary data examples, showing participants how to work their way through the analysis, and answering questions about the data.
Class sessions:
- Introduction to R and reproducibility (including using knitr to incorporate code into pdf, html, Word documents)
- Introduction to plotting data (mapping using ggplot2, scatterplots, bar charts, time series, profiles, boxplots)
- Rearranging and cleaning data (learn to use packages tidyr, dplyr, and lubridate)
- Advanced plotting and polishing graphics (using cognitive principles for publication graphics; packages ggplot2, maps, ggmap, ggsubplot, productplots, GGally, gridExtra)
- Making shiny apps (create interactive web applets for users to interact with their/your/public data and analyses)
2015 Annual Business Meeting
Keynote Speaker: Barbara Bailey, Associate Professor of Statistics, San Diego State University
Nonlinear Models and Prediction Intervals for Plankton Ecosystem Dynamics
Also Included in this year’s Business Meeting: Dinner, Chapter news updates
Time series of physical and biological properties of the ocean are a valuable resource for developing models for ecological forecasting and ecosystem-based management. We describe the development of a nonlinear model that predicts the abundance of the important zooplankton species Calanus finmarchicus from hydrographic data from the Gulf of Maine. Forecasts are constructed for the model fit to 1978-2003 bimonthly data and corresponding forecasts intervals are obtained by the stationary bootstrap. The properties of the distribution free prediction intervals based on the stationary bootstrap will be investigated. The results of neural network model fits, forecasts, and forecast intervals to nonlinear time series data generated from the Rossler nonlinear dynamical system will be presented.
Twenty people attended. Slides can be found here.
2014 Fall Roundtable
The fall roundtable was held on 121/2014 at SDSU. The topic was "When is Big Data Useful?", and the speakers were Dr. Andrew Baker (SDSU) and Dr. Alexander Kolovos (Teradata). Twenty people attended.
2014 Spring Roundtable
The spring roundtable was held on 4/22/14 at Pfizer Inc. The speaker was Dr Yoav Freund, professor of Computer Science and Engineering at UCSD. He spoke on Big Data. Thirty people attended.
Abstract
At the beginning of the 90's, data sets with 10,000 instances and 100 dimensions were considered "big". In 2014, data sets with a million examples and a million dimensions (sparsely populated) are common. This massive change in scale requires a reformulation of the basic tools of data analysis. On the data management side we have hdfs/hadoop, column based databases and distributed databases. On the statistical analysis level we have stochastic gradient descent, support vector machines, boosting and compressed sensing. Much progress has taken place on each side of the statistics / data management fence but integration of the two sides is still lagging. In this talk I will briefly review the current state of affairs and describe some of the work taking place in UCSD.
2014 Annual Business Meeting
The Annual Business meeting was held on 2/6/14. 20 people attended. SDASA presenters were the president, Jihao Jhou, with a review of 2013 events and upcoming events for 2014, and the treasurer, Patricia English, with a review of 2013 finances. In addition, Dr Chiu-Hsieh "Paul" Hsu, Associate Professor, Mel and Enid Zuckerman College of Public Health, University of Arizona, spoke on Multiple Imputation Using Smoothing Techniques. The slides are available here.
The 2013 SDASA One-Day Conference!
On Oct 16, 2013, the SDASA was pleased to hold a one-day conference. It was held at Pfizer La Jolla, and ran from 8:30 am to 3:30 pm. 44 people attended.
Our Keynote speaker was Dr. Nick Jewell, Professor of Biostatistics, from the UC Berkeley School of Public Health. He spoke on Trajectory Modeling By Shape (longitudinal data analysis yielding clusters.)
We also invited 6 local speakers to give talks. They were (alphabetical):
Wei Cheng, ISIS Pharmaceuticals: Taking the Leap - Using ODS Statistical Graphics
Photos (clockwise from left):
Drs. Nick Jewell and Zihao Zhou, Drs Jihao Zhou, Nick Jewell, Larry Shen, John Chalekian,
Students enjoying the talks, Dr. Lily Xu
Annual Picnic
The annual picnic we held on September 22, 2013, from 11:30 am - 2 pm at the Ocean Air Community Park. Food was provided by Green Leaf Cafe & Catering. Approximately 25 people attended and had a great time.
Ocean Air Community Park opened on January 20, 2010. It is an 18 acre park that includes a 16,500 square foot Recreation Center providing meeting rooms, a kitchen, offices, restrooms, and a multi-functional gymnasium. The park includes two ball fields, an outdoor basketball court, walking paths, picnic structures, and a large grassy field surrounded by rolling mounds. The park also provides a large playground with climbing rocks and boulders, a sand area, and various play structures for children ages 2 through 12.
Artistically, the park includes a long row of Liquidambar trees that, which when mature will shield park users from prevailing winds that sweep through the site. The site includes California native plants that line the pathways and bring the users up to a stainless steel deck that overlooks the open space preserve. The project was designed to meet the Leadership in Energy and Environmental Design (LEED) silver certification for its environmentally sustainable design.
Spring Roundtable.
The Spring Roundtable was held on May 30, 2013, at Pfizer La Jolla. There were 15 attendees. We had two speakers.
Shibing Deng, Associate Director Statistics, Pfizer Worldwide Research and Development, La Jolla, CA, spoke on Identifying Outliers in Oncology Gene Expression Studies. His slides are available here.
Cheryl Rossi, VP Biostatistics and Epidemiology, BioRxConsult Inc, spoke on Handling Treatment Heterogenity (HTE): Overview and Methods. Her slides are available here.
On February 24, 2013, we held our Annual Business Meeting at Balboa Park. The agenda included two guest speakers. Bob Newcomb, spoke on The Past, Present and Future of the San Diego Chapter of the ASA (he helped found the chapter in 1999), and Bob Riffenburgh, who presented results from his research: On Assuming Normality for a t Test. In addition, there was a summary of 2012's financial status by our Treasurer, Patricia English, the 2012 Council of Chapters meeting at JSM was presented by our Chapter Representative, Matt Marler, and a review of upcoming 2013 events by our President, Colleen Kelly. Lunch was served. 15 people attended.
On October 18, 2012, we held our Fall Statistics Roundtable at Pfizer La Jolla. The topic was Missing Data Issues in Statistics Analysis. Professor Fang Liu, Department of Statistics, Notre Dame University. Approximately 25 people attended.
On October 6, 2012, we held our annual picnic at Crown Point North, Mission Beach. Around
30 people attended.
On September 11, 2012, Pfizer La Jolla hosted the ASA travelling course "Workshop on Applied Survival Analysis." The speaker was Susanne May, Ph.D., University of Washington, Seattle. Approximately 60 people attended and enjoyed the workshop. If you wish to see what you missed, check the course description at
the ASA website: http://www.amstat.org/education/travelingcourses/tc_sa.cfm
Adaptive Design Roundtable
Professor Yisheng Li, U of Texas & MD Anderson Cancer Center, at Amylin Pharmaceuticals
2011 Picnic
2001-2005 Events
2001-2005 Events