Exploring Survival Analysis Designs for Clinical Trials
- Piecewise Weighted Linear Regression.nqt
- MaxCombo.nqt
- Generalized Piecewise Weighted Linear Regression.nqt
- Two Sample Linear Regression.nqt
- Linear Regression Simulation.nqt
Exploring Survival Analysis Designs for Clinical Trials
Identifying & Addressing Challenges With Survival Sample Size & Power
Survival analysis is one of the most common statistical approaches used in clinical trials, especially in clinical areas such as oncology.
However, it is also one of the most complex and flexible statistical areas with this increasing in recent years due to the emergence of innovative treatments such as immunotherapies.
This complexity affects every part of trial design and analysis including power analysis and sample size determination.
You will learn about:
- Power Calculations for Survival Analysis
- Inputs Required for Survival Power Calculations
- Issues & Challenges for Survival Power Calculations Demonstration Including Worked Examples for:
- Log-Rank Test
- Linear-Rank Tests (FlemingHarrington & Modestly Weighted Tests)
- MaxCombo Procedure
Exploring Survival Analysis Designs for Clinical Trials
A key difference between clinical trials with survival endpoints and other endpoints (such as comparing means and proportions) is that in a survival analysis, the power is directly linked to the number of events and the researcher is recruiting the sample size they expect is required to achieve a target number of events.
In practical terms, this means there is significant flexibility required to deal with issues such as varying accrual, follow-up length, hazard rate and dropout patterns even within the “standard” two independent group log-rank scenario.
Another design issue to consider is whether all subjects should be followed for the same fixed period of time or if each subject should be observed until the end of the study once they are recruited.
These considerations will impact the total sample size or overall trial length required to achieve the target number of events.
In addition to this, the issue of non-proportional hazards (where the hazard ratio varies over time) has been of particular interest in recent years, especially in the context of immunotherapies which commonly have a “delayed effect”.
Multiple proposals have emerged for the analysis of these complex survival datasets such as weighted linear rank tests (e.g. Fleming-Harrington, Modestly Weighted) and the MaxCombo test.
In this tutorial, we have delved into survival analysis within the context of clinical trials and scrutinised various challenges researchers may encounter.
nQuery helps make your clinical trials faster, less costly and more successful with tools for sample size calculation - refining Frequentist, Bayesian & Adaptive designs.
Start your unlimited 14-day trial to help design efficient, informative and ethical adaptive trials.
Who is this for?
This will be highly beneficial if you're a biostatistician, scientist, or clinical trial professional that is involved in sample size calculation and the optimization of clinical trials in:
- Pharma and Biotech
- CROs
- Med Device
- Research Institutes
- Regulatory Bodies