Constructing an External Control Arm Using the FrenchCohort to Replicate Outcomes of a Randomized Trial Arm in Advanced Hodgkin Lymphoma
Source: PubMed Central Open Access, NCBI / U.S. National Library of Medicine
External control (EC) arms using routine data can be used as comparators in clinical trials under certain conditions. The objective of this study was to assess the ability of an EC arm to replicate outcomes of a randomized controlled trial (RCT) arm receiving the same treatment strategy in advanced stage classical Hodgkin lymphoma and serve as comparator in future trials. Using the prospective real‐world data in lymphoma and survival in adults (REALYSA) cohort, we constructed an EC arm by applying eligibility criteria of the AHL2011 phase III RCT. Propensity score (PS) weighting was used to balance baseline characteristics between the EC arm and the RCT arm that received the same treatment. Progression‐free survival (PFS) and overall response rate (ORR) at the end of first‐line treatment were estimated in both arms. Hazard ratio (HR) for PFS and odds ratio (OR) for ORR were estimated using the EC arm as reference. A sensitivity analysis evaluated PS matching. Four hundred and ten RCT patients and 191 EC patients were analyzed. After PS weighting, patients' characteristics were well‐balanced between arms. PFS estimates were similar in RCT and EC arms with an HR close to 1 (HR, 0.98; 95% CI, 0.59–1.62). ORR estimates were comparable in both arms. Weighting performed better than matching. Using PS weighting gave similar outcomes in the EC arm compared with the RCT arm. Our results suggest that EC studies using routine care data could be considered in the development of strategie
Abstract
External control (EC) arms using routine data can be used as comparators in clinical trials under certain conditions. The objective of this study was to assess the ability of an EC arm to replicate outcomes of a randomized controlled trial (RCT) arm receiving the same treatment strategy in advanced stage classical Hodgkin lymphoma and serve as comparator in future trials. Using the prospective real‐world data in lymphoma and survival in adults (REALYSA) cohort, we constructed an EC arm by applying eligibility criteria of the AHL2011 phase III RCT. Propensity score (PS) weighting was used to balance baseline characteristics between the EC arm and the RCT arm that received the same treatment. Progression‐free survival (PFS) and overall response rate (ORR) at the end of first‐line treatment were estimated in both arms. Hazard ratio (HR) for PFS and odds ratio (OR) for ORR were estimated using the EC arm as reference. A sensitivity analysis evaluated PS matching. Four hundred and ten RCT patients and 191 EC patients were analyzed. After PS weighting, patients' characteristics were well‐balanced between arms. PFS estimates were similar in RCT and EC arms with an HR close to 1 (HR, 0.98; 95% CI, 0.59–1.62). ORR estimates were comparable in both arms. Weighting performed better than matching. Using PS weighting gave similar outcomes in the EC arm compared with the RCT arm. Our results suggest that EC studies using routine care data could be considered in the development of strategies for advanced Hodgkin lymphoma.