Substituting a randomised placebo control group with a historical placebo control in an endometriosis pain trial: a case study re-evaluating trial data using historical control data from another trial

Abstract

ObjectiveThe substitution of an in-study control population with a historical control (HC) population is considered a viable option for reducing the necessary recruitment of control patients. However, it is necessary to evaluate whether this method is applicable to studies on indications targeting endometriosis-associated pelvic pain (EAPP). This study aims to evaluate the potential bias in the results of an EAPP study with an HC arm.MethodsFor this case study, we re-evaluated data from a randomised, placebo-controlled trial using dienogest daily to treat EAPP with an HC arm based on data from a second randomised, placebo-controlled trial in the same indication. Propensity Score (PS) matching was used to match between the treatment and HC arm on all baseline variables. To evaluate the effect of matching on the introduced bias, we evaluated efficacy parameters with the full treatment and control group, as well as the matched group.ResultsThe difference between means (placebo minus treatment) in change in pain, as measured on the Visual Analogue Scale from baseline to end of treatment, deviates in the comparison treatment/pool of HC (7.15 (0.22 to 14.08)) from the overall in-study group (reference: 11.89 (6.06 to 17.73)). After PS matching on the baseline variables, the difference between means (11.79 (4.09 to 19.5)) is close to the reference.ConclusionsUsing HC with PS matching has proven to be useful in the setting of treating EAPP, while emphasis must be given to the selection mechanism and the underlying assumptions. This case study has shown that even for studies which are very similar in design, heterogeneity and between-study variations are present. With the use of an HC arm, it was possible to reproduce similar results than in the original study, while the PS matching improved the comparability considerably. For the main endpoint, PS matching could reproduce the original study results.Trial registration numberNCT00225199,NCT00185341