Background: Real-world evidence (RWE) analytics continue to advance natural history comparisons in rare diseases. The Global Alagille Alliance (GALA) is the largest global clinical research database for Alagille syndrome (ALGS). Maralixibat (MRX) is an ileal bile acid transporter inhibitor approved by the FDA for the treatment of cholestatic pruritus in patients with ALGS ≥1 year of age. A pre-specified analysis plan applied novel analytical techniques to compare RWE from GALA with a MRX cohort with the aim to compare event-free survival (EFS) in patients with ALGS.

Methods: GALA contains retrospective data for clinical parameters, biochemistries and outcomes. The MRX database comprises of 84 ALGS patients with up to 6 years of data. EFS was defined as the time to first event of hepatic decompensation (variceal bleeding, ascites requiring therapy), surgical biliary diversion, liver transplantation (LT), or death. GALA was filtered to align key MRX eligibility criteria. The index time was determined via maximum likelihood estimation. Balance among baseline (BL) variables was assessed. Selection of patients and index time was blinded to clinical outcomes. Sensitivity and subgroup analyses, and adjustments for covariates, were applied. Missing outcomes data were censored at last contact.

Results: Of 1,438 patients in GALA, 469 were eligible. Age, total bilirubin (TB), gamma glutamyltransferase (GGT) and alanine aminotransferase (ALT) were well balanced between groups and no statistical differences were observed for age, mutation, region, TB, GGT and ALT. Median BL serum bile acids (sBA) was significantly higher in the MRX cohort (p=0.003); 85% of sBA data was not available in GALA. EFS rates in the MRX cohort were significantly better than those reported in the GALA control, (crude 6-year EFS: 73% and 50%, respectively, and adjusted for age, sex, TB, ALT: HR=0.305; 95% CI: 0.189–0.491; p<0.0001). Varied index times, weighted inverse probability of treatment weights, average treatment effect in the treated, LT and death only, regions, sBA sub-group, pruning events to 12 months were consistent with the primary result. Limitations include no standardised measure of pruritus and limited sBA data in GALA, and inherent bias for patients who enter a clinical trial.

Conclusions: This 6-year analysis suggests the potential for improved EFS with MRX in patients with ALGS. This RWE analysis provides a potential method to evaluate outcomes in long-term intervention studies where placebo comparisons are not feasible. Limitations will always be present given lack of prospective conduct and inherent biases, though sensitivity analyses can help mitigate and aid interpretation.