Pre-transplant Biomarkers of Immune Dysfunction Improve Risk Assessment of Post-transplant Mortality Compared to Conventional Clinical Risk Scores.

Introduction: There is a critical need to accurately stratify liver transplant (LT) candidates' risk of post-LT mortality prior to LT to optimize patient selection and avoid futility. Here, we compare previously described pre-LT clinical risk scores with the recently developed Liver Immune Frailty Index (LIFI) for prediction of post-LT mortality. LIFI measures immune dysregulation based on pre-LT plasma HCV IgG, MMP3 and Fractalkine. LIFI accurately predicts post-LT mortality, with LIFI-low corresponding to 1.4% 1-year post-LT mortality compared with 58.3% for LIFI-high (C-statistic=0.85). Methods: LIFI was compared to MELD, MELD-Na, MELD 3.0, D-MELD, MELD-GRAIL, MELD-GRAIL-Na, UCLA-FRS, BAR, SOFT, P-SOFT, and LDRI scores on 289 LT recipients based on waitlist data at the time of LT. Survival, hazard of early post-LT death, and discrimination power (C-statistic) were assessed. Results: LIFI showed superior discrimination (highest C-statistic) for post-LT mortality when compared to all other risk scores, irrespective of biologic MELD. On univariate analysis, the LIFI showed a significant correlation with mortality 6-months, as well as 1-, 3-, and 5-years. No other pre-LT scoring system significantly correlated with post-LT mortality. On bivariate adjusted analysis, African American race (p<0.05) and pre-LT cardiovascular disease (p=0.053) were associated with early- and long-term post-LT mortality. Patients who died within 1-yr following LT had a significantly higher incidence of infections, including 30-day and 90-day incidence of any infection, pneumonia, abdominal infections, and UTI (p<0.05). Conclusions: LIFI, which measures pre-LT biomarkers of immune dysfunction, more accurately predicts risk of post-LT futility compared with current clinical predictive models. Pre-LT assessment of immune dysregulation may be critical in predicting mortality after LT and may optimize selection of candidates with lowest risk of futile outcomes.