Jennings, Linda W.

Abstract Context. —Establishing adequate interobserver agreement is crucial not only for standardization of patient care but also to ensure validity of findings in multi-institutional trials. Objective. —To evaluate interobserver agreement in assessing chronic hepatitis C (HCV) and acute cellular rejection (ACR) among 17 hepatopathologists involved in the “Hepatitis C 3” trial. Design. —The trial is a randomized multicenter (17 institutions) study involving 312 patients undergoing transplantation for HCV. Patients are randomized to 3 treatment arms. For final data analysis, all biopsy specimens are reviewed by a central pathologist (G.J.N.). Recurrence of HCV is evaluated according to the Batts and Ludwig schema. The 1997 Banff schema is used to evaluate ACR. To assess interobserver agreement, hematoxylin-eosin–stained sections from 11 liver biopsy specimens (6 HCV and 5 ACR) were sent by the central pathologist to 16 local pathologists from 13 institutions. Statistical analysis was performed on raw ACR/HCV data as well as data grouped according to clinically significant primary endpoint cutoffs. Results. —Statistically significant agreement was found among all participating pathologists ( P < .001). On κ analysis, the degree of agreement was rated “moderate” for HCV grade and stage and ACR global grading (κ = 0.30, 0.33, and 0.37, respectively). Interobserver agreement was weaker for rejection activity index scoring of ACR (κ = 0.15). A stronger degree of agreement was found when scores were grouped based on endpoint cutoffs (κ = 0.76 “almost perfect” for HCV and 0.62 “substantial” for ACR). Conclusions. —An overall statistically significant interobserver agreement was found among 17 pathologists using the 1997 Banff schema and the Batts and Ludwig schema.

Background and Aims Among patients with cirrhosis awaiting liver transplantation, prediction of wait‐list (WL) mortality is adjudicated by the Model for End Stage Liver Disease–Sodium (MELD‐Na) score. Replacing serum creatinine (SCr) with estimated glomerular filtration rate (eGFR) in the MELD‐Na score may improve prediction of WL mortality, especially for women and highest disease severity. Approach and Results We developed (2014) and validated (2015) a model incorporating eGFR using national data (n = 17,095) to predict WL mortality. Glomerular filtration rate (GFR) was estimated using the GFR assessment in liver disease (GRAIL) developed among patients with cirrhosis. Multivariate Cox proportional hazard analysis models were used to compare the predicted 90‐day WL mortality between MELD‐GRAIL‐Na (re‐estimated bilirubin, international normalized ratio [INR], sodium, and GRAIL) versus MELD‐Na. Within 3 months, 27.8% were transplanted, 4.3% died on the WL, and 4.7% were delisted for other reasons. GFR as estimated by GRAIL (hazard ratio [HR] 0.382, 95% confidence interval [CI] 0.344‐0.424) and the re‐estimated model MELD‐GRAIL‐Na (HR 1.212, 95% CI 1.199‐1.224) were significant predictors of mortality or being delisted on the WL within 3 months. MELD‐GRAIL‐Na was a better predictor of observed mortality at highest deciles of disease severity (≥ 27‐40). For a score of 32 or higher (observed mortality 0.68), predicted mortality was 0.67 (MELD‐GRAIL‐Na) and 0.51 (MELD‐Na). For women, a score of 32 or higher (observed mortality 0.67), the predicted mortality was 0.69 (MELD‐GRAIL‐Na) and 0.55 (MELD‐Na). In 2015, use of MELD‐GRAIL‐Na as compared with MELD‐Na resulted in reclassification of 16.7% (n = 672) of patients on the WL. Conclusion Incorporation of eGFR likely captures true GFR better than SCr, especially among women. Incorporation of MELD‐GRAIL‐Na instead of MELD‐Na may affect outcomes for 12%‐17% awaiting transplant and affect organ allocation.

Background and Aims A high proportion of patients develop chronic kidney disease (CKD) after liver transplantation (LT). We aimed to develop clinical/protein models to predict future glomerular filtration rate (GFR) deterioration in this population. Approach and Results In independent multicenter discovery (CTOT14) and single‐center validation (BUMC) cohorts, we analyzed kidney injury proteins in serum/plasma samples at month 3 after LT in recipients with preserved GFR who demonstrated subsequent GFR deterioration versus preservation by year 1 and year 5 in the BUMC cohort. In CTOT14, we also examined correlations between serial protein levels and GFR over the first year. A month 3 predictive model was constructed from clinical and protein level variables using the CTOT14 cohort (n = 60). Levels of β‐2 microglobulin and CD40 antigen and presence of hepatitis C virus (HCV) infection predicted early (year 1) GFR deterioration (area under the curve [AUC], 0.814). We observed excellent validation of this model (AUC, 0.801) in the BUMC cohort (n = 50) who had both early and late (year 5) GFR deterioration. At an optimal threshold, the model had the following performance characteristics in CTOT14 and BUMC, respectively: accuracy (0.75, 0.8), sensitivity (0.71, 0.67), specificity (0.78, 0.88), positive predictive value (0.74, 0.75), and negative predictive value (0.76, 0.82). In the serial CTOT14 analysis, several proteins, including β‐2 microglobulin and CD40, correlated with GFR changes over the first year. Conclusions We have validated a clinical/protein model (PRESERVE) that early after LT can predict future renal deterioration versus preservation with high accuracy. This model may help select recipients at higher risk for subsequent CKD for early, proactive renal sparing strategies.