Prospective evaluation of vibration-controlled transient elastography for guiding long-term protocol liver biopsies after liver transplantation

• Cohort of 90 patients between 5 and 10 years post liver transplantation with PBH and VCTE • Histological and VCTE quality criteria • No missing data on the tests studied • Histological analysis performed by an expert pathologist • Blind reclassification of biopsies by two expert hepatologists • First study comparing all protocol liver biopsies to non-invasive tests after 5 years post liver transplantation • Good acceptability and feasibility of VCTE in this population (3.6% failure rate). • Good diagnostic performance of VCTE to predict abnormal PBH (AUROC: 0.816). • Elasticity threshold with the best performance at 6.90 kPa. • CAP ≥285 dB/m accurately detected significant steatosis in this population. • VCTE could help avoid systematic protocol liver biopsy and select patients at risk of developing graft disease after 5 years post-transplant. Liver biopsy (LB) is the gold standard for diagnosing post-liver transplantation (LT) complications. Currently, there is no data on the value of vibration-controlled-transient elastography (VCTE) for medium- and long-term follow-up after LT compared to protocol LB in asymptomatic patients. This study aims to evaluate the correlation between VCTE-derived liver stiffness measurements (LSM) and histological abnormalities observed in protocol LB 5 to 10 years after LT, to determine the potential role of VCTE in long-term graft monitoring. A prospective study was conducted at a liver transplantation center in Montpellier, France, involving adult LT recipients who underwent both VCTE and LB within 6 months between January 2022 and January 2024. Protocol LB were performed in the absence of clinical or laboratory evidence of post-LT complications. The primary endpoint was to determine the LSM threshold for predicting abnormal LB. Receiver operating characteristic (ROC) curve analysis and the Youden index were used to identify the optimal cutoff. Among 90 patients, 32% had abnormal LB findings, including fibrosis ≥ F1 (by Ishak) in 58.6% of abnormal biopsies. The mean LSM was significantly higher in patients with abnormal LB (12.6 ± 13.9 kPa) compared to those with normal LB (5.9 ±1.6kPa), p < 0.001. LSM (area under the ROC curve [AUC]: 0.816) was the most accurate predictor of abnormal LB. The optimal LSM threshold to predict abnormal LB was 6.90 kPa (sensitivity [Se] = 0.69, specificity [Sp] = 0.84), while values ≤5.6 kPa and >12.9 kPa respectively ruled out and predicted an abnormal LB. These results underscore the relevance of VCTE in long-term follow-up of LT recipients to determine the need for LB. VCTE effectively predicts abnormal histology and can guide selective use of protocol liver biopsies, potentially reducing unnecessary procedures. Multicenter studies are needed to validate these findings.