A novel approach to augment single-arm clinical studies with real-world data

Using real‐world data to predict health outcomes—The prediction design: Application and sample size planning

“…In the literature, it is argued that the requirement of classical RCTs, that is, patients being similar and groups being homogeneous, is incompatible with the molecular diversity and diverse therapeutic options of the future personalized medicine approaches (Klauschen et al., 2014) and that it therefore seems “[…] likely that there will be shifts in the judgment of the scientific community whether RCTs are the appropriate standard for development of drugs for diseases with high patient heterogeneity, once the big data concept has effectively entered practice” (Eichler et al., 2016). On the other hand, it is important to take into account possible limitations of RWD sources, as, for example, inconsistencies in the data (e.g., related to study duration, assessment frequency, and outcome measurements), problems concerning the integration of data between different systems, and lack of important covariates (Jiao et al., 2022; Sachdeva et al., 2021). Overall, it must be critically discussed whether the data are sufficient to address the causal question of interest (Burger et al., 2021; Campbell, 2021; Hampson et al., 2021; Levenson et al., 2021) and if real‐world evidence can be understood as “substantial evidence” in regulatory decisions (Song et al., 2021).…”

“…Sachdeva et al. (2021) and Yin et al. (2022) proposed methods to augment single‐arm clinical trials with external data sources, where they use the propensity score methodology to achieve covariate balance between the groups to carry out inference for the outcome variable.…”

“…(2022) is a case study in non–small cell lung cancer, Sachdeva et al. (2021) investigated type I and type II error rates depending on different values for the sample size in the single‐arm trial. However, they do not present a closed‐form solution sample size formula.…”

Navigating the Budding Researchers through Different Study Designs in Homeopathy: Part 1 of Research Method Series

“…They are often used in situations where a control group is not feasible or ethical. 26 The primary objective of a single-arm trial is to evaluate the safety and efficacy of a new intervention or treatment. Single-arm trials can also provide information on feasibility and acceptability of the intervention, as well as identify potential side effects or complications.…”

A Propensity-Score Integrated Approach to Bayesian Dynamic Power Prior Borrowing

“…However, a considerable difference may still exist and lead to a moderate bias. To mitigate the impact of population difference, some recently developed approaches used PS matching or stratification to reduce the difference within matched pairs or strata, and then applied the power prior within them (Wang et al, 2019a(Wang et al, , 2019bSachdeva et al, 2021). An alternative approach used a linear outcome model assuming exchangeability after covariate adjustment (Kotalik et al, 2021).…”