(H) Concordance of Efficacy Data Collected in a Clinical Trial Setting Versus a Registry Study Setting: Example of Use of Real-World Data to Support Drug Regulatory Approval of Abatacept for Acute Graft Versus Host Disease Prophylaxis

Publication Number: 666

Background: Abatacept was approved for acute graft versus host disease (aGVHD) prophylaxis in recipients of unrelated-donor hematopoietic stem cell transplantation (HSCT) in Dec 2021. This is the first regulatory approval for aGVHD prophylaxis, and one that incorporated real-world evidence (RWE) in the assessment of clinical efficacy. For this approval, findings from 2 clinical studies were reviewed: a multicenter phase 2 clinical trial (1) (ABA2; NCT01743131) and a prospectively designed, retrospective review of registry data from the Center for International Blood and Marrow Transplant Research (CIBMTR). In the US, all data on patients undergoing allogeneic HSCT are routinely reported by centers and entered into the CIBMTR database. For regulatory review, the accuracy of data collected in the CIBMTR database was assessed in comparison with data from the ABA2 trial.

Objectives: To assess the measurement bias/degree of concordance of patient-level data from ABA2 trial participants with their data collected in the CIBMTR database.

Methods: This descriptive study compared data from HSCT recipients who enrolled in the ABA2 trial with data in the CIBMTR database for concordance. Patients were stratified into 2 cohorts: 7/8 human leukocyte antigen (HLA)-matched (n = 42) or 8/8 HLA-matched unrelated-donor transplant recipients (n = 137). The primary outcome (overall survival), baseline covariates, independent variables, treatment exposure, and other outcome endpoints were evaluated.

Results: Among the 7/8 and 8/8 HLA-matched patients, there was overall consistency between data reported in the ABA2 trial and the CIBMTR database. In the 7/8 HLA-matched cohort, 100% agreement was found in GVHD prophylactic regimens, sex, conditioning regimen, stem cell source, and year of transplant; there was almost 100% agreement in age. Additionally, 100% agreement was observed between data sources for the primary outcome of death and the combined outcome of relapse or death up to day 100 or 180. For the combined outcome of grade 2–4 aGVHD or death up to day 180, there was 95.2% agreement between the two data sources. In the 8/8 HLA-matched patients, death up to day 180 showed 99.3% agreement. Further, 100% agreement was observed in GVHD prophylactic regimens, sex, conditioning regimen, and year of transplant.

Conclusions: A high degree of concordance was noted in patient information and outcome variables between the ABA2 data and CIBMTR database, which reflects the high quality of information collected by the CIBMTR. Regulatory submission of data from the ABA2 clinical trial and RWE from the CIBMTR database resulted in the recent approval of abatacept for aGVHD prophylaxis; this is the first and only therapy approved to date for this indication.

1. Watkins B, et al. J Clin Oncol 2021;39:1865–77.