Synthetic control arm (SCA) methods are statistical methods that are seeing rapidly increasing use in comparative effectiveness research. SCA analyses often involve comparing single-arm trials against external control arms constructed from real-world data (RWD) where conducting randomized clinical trials is difficult or infeasible. These benefits are especially evident in rare disease trials where sample sizes are typically substantially smaller, and it is difficult to determine standard-of-care treatments. Rare disease trials conducted with very few patients translates to insufficient statistical power or are performed as single-arm trials that make it difficult to compare against other therapeutic options without SCA methods. Advanced statistical methods are applied to RWD to build the SCA in a way that allows for the appropriate comparison with data gathered during the execution of the single arm trial.
However, each synthetic control project has its own unique challenges with regards to generalizability of results, interpretation and associated statistical methodology. In rare disease indications, the severely limited sample sizes in both single-arm trial and RWD, can present design challenges for SCAs. At ISPOR US 2022, Eric Mackay and Aaron Springford, Research Principal at Cytel, contributed a podium presentation on ‘Power Implications of Estimator Choice in Synthetic Control Arm Analyses’.
SCA Design Challenges
The right approach to designing an SCA depends on the types of data available to sponsors, the desired sample size, and the anticipated length of a trial. Once the appropriate data sources are identified, a careful assessment needs to be made about the potential biases which are inherent in RWD or present in historical trial data as they were not collected for the purposes of the comparison with a new single arm trial.
Unlike in the experimental setting of the randomized clinical trial, the likelihood of selection bias is higher when using real world data. Such bias occurs when the patients receiving the experimental treatment differ, in terms of their baseline prognostic characteristics, from those receiving comparator treatments in a real-world setting.
Propensity Scoring Methods for SCA
The propensity score, as defined by Rosenbaum & Rubin, refers to the probability of receiving a given treatment. Propensity score adjustment is a powerful set of statistical techniques that can mitigate bias due to observed confounders when performing indirect treatment comparisons. The way to adjust for bias varies, but a common way is to match individuals in the treated and untreated cohorts, who have identical propensity scores and then compare differences in outcomes between pairs. Other ways to adjust are to compare outcomes within propensity score strata; or to compare outcomes after applying weights inversely proportional to each patient’s propensity score.
Practitioners often estimate an average treatment effect (ATE) or average treatment effect on the treated (ATT) via inverse probability of treatment weighting. However, ATT may not always maximize power and, depending on cohort relative sample sizes, estimators like average treatment effect on the untreated (ATU) could outperform. In his presentation, Eric will illustrate the power implications of choice of ATT vs. ATU when performing an indirect treatment comparison using propensity score weighting. Power will be computed across a range of simulation scenarios reflective of the settings in which SCAs are often performed.
Another important podium presentation from Cytel, ‘Studies on COVID-19 Healthcare Impacts’ is going to be led by Cytel’s Senior Research Consultant, Rachel Knapp. Rachel will present a study that sought to assess the impact of the COVID-19 pandemic and resultant lockdown measures on the staging of incident breast cancer diagnoses in the United States, using data from 2018-2021.
Click below to download Cytel's full list of sessions at ISPOR US 2022.
Cytel will be at ISPOR in person in 2022. Please visit us at Booth #725.
About the Author of Blog:
Mansha Sachdev specializes in content creation and knowledge management. She holds an MBA degree and has over 12 years of experience in handling various facets of marketing, across industries. At Cytel, Mansha is a Senior Content Marketing Manager and is responsible for producing informative content that is related to the pharmaceutical and medical devices industries.