GLOBE SCORE PERFORMANCE IN PREDICTING URSODEOXYCHOLIC ACID TREATMENT RESPONSE IN A SINGLE UNITED STATES COHORT

Background
Effective primary biliary cholangitis (PBC) management is vital due to its progression to cirrhosis and its impact on survival. Ursodeoxycholic acid (UDCA) is the main therapy, but predicting treatment response is challenging. The GLOBE score is a validated predictive model for liver transplant (LT)-free survival in PBC patients treated with UDCA. However, its ability to predict UDCA treatment response remains inadequately explored. This study aims to evaluate the GLOBE score's effectiveness in predicting treatment response among PBC patients undergoing UDCA therapy, shedding light on factors associated with treatment outcomes.

Methods
In a year-long prospective observational cohort study at a U.S. liver center, we monitored UDCA-treated PBC patients. Exclusions covered those with additional chronic liver diseases, participants with PBC/autoimmune hepatitis (AIH) overlap syndrome, and individuals inconsistently treated with UDCA or who discontinued treatment. Our sample included pure PBC patients, with responders defined by a GLOBE score ≤0.30. We assessed the predictive efficacy of the GLOBE score for UDCA treatment response, comparing it to the Paris-2 criteria. Subsequently, univariate and multivariate analyses were conducted, adjusting for demographic and biochemical variables to assess comprehensively the score's efficacy.

Results
We evaluated 136 PBC patients under continuous UDCA therapy for a year. Most were female (90.2%) and predominantly white caucasian (79.0%), with an average age of 56 years. For diagnosing PBC, 82% of patients met ALP criteria, 63% had positive antibodies, and 51% had compatible liver biopsies. Liver function markers at UDCA initiation and after 12 months showed notable differences. Initially, ALP levels were higher (221 vs 173), and both ALT (55 vs 33) and AST (47 vs 33) levels exhibited a notable decline. At end of follow-up, responder PBC patients exhibited a mean GLOBE of -0.56 (SD = 1.02). GLOBE Score performance, compared to Paris-2 criteria, indicated a non-responder rate of 17% (p = 0.2050) and 19% (p = 0.014), respectively. The Cox regression multivariate analysis, adjusted for age, gender, and biochemical markers, revealed a robust association of the GLOBE score among PBC responders to UDCA treatment (adjusted odds ratio 9.30, 95% confidence interval 3.2-16.9, p < 0.0001).

Conclusion
Our study highlights the potential of the GLOBE score in predicting UDCA treatment response for PBC patients. While we acknowledge limitations, such as the difficulty in calculating LT-free survival, these challenges emphasize the importance of extended observation periods for a more comprehensive understanding. While existing data supports the GLOBE score as a survival predictor, its validation as a treatment response predictor necessitates further research with larger cohorts and extended follow-up durations.