gms | German Medical Science

Background: The interpretation of intention-to-treat analyses of randomised clinical trials is often hindered as a result of noncompliance and treatment switching. This has recently given rise to a vigorous research activity on the identification and estimation of so-called estimands, which can bring a more refined insight into the treatment effect.

We aim to estimate the treatment effect in the so-called switchers in a randomised clinical trial (e.g., patients who were switched to a different dose or treatment). This effect expresses how different the average outcome would have been for them, had they not switched.

Methods: Motivated by an ongoing clinical trial conducted by Janssen Pharmaceuticals in which a flexible dosing regimen is compared to placebo, we evaluate how switchers in the treatment arm (i.e., patients who were switched to the higher dose) would have fared had they been kept on the low dose. To realize this, we will transport data from a fixed dosing trial that has been conducted concurrently on the same target, albeit not in an identical patient population. We formulate the statistical problem in terms of potential outcomes and discuss the plausibility of the assumptions required to transport the mean of the potential outcomes under a given treatment regimen from one trial to another. To avoid undue reliance on modelling assumptions, we propose a doubly robust estimator, which relies on an outcome model and a propensity score model for the association between study and patient characteristics but only requires one of them to be correctly specified. We discuss practical considerations regarding variable selection and model specification.

Results: The proposed estimator is easy to evaluate, asymptotically unbiased if either model is correctly specified and efficient (under the model defined by the restrictions on the propensity score) when both models are correctly specified. Monte Carlo simulations and application to a clinical trial conducted by Janssen demonstrated adequate performance.

Conclusion: Causal inference techniques on transportability made it possible to estimate treatment effects for switchers in randomised clinical trials.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.