About the LEOPARD project
Overview
The LEOPARD project represents a collaborative initiative by European liver transplantation (LT) stakeholders dedicated to modernising and refining organ allocation strategies for decompensated cirrhosis (DC) and hepatocellular carcinoma (HCC) patients.
While LT represents a life-saving procedure for these patients, its efficacy is hindered by a persisting risk of death or dropout while on the waitlist, primarily due to organ shortage.
To address this challenge, organ sharing organisations (OSOs) prioritise LT candidates based on the severity of their condition using predictive models such as the Model for End-Stage Liver Disease (MELD) which predicts mortality in DC but not in HCC cases. However, with a rising number of HCC candidates and other major changes in the profile of LT candidates, this model encounters increasing limitations, resulting in disparities in LT access, and persisting mortality rates ranging from 15 to 30% in countries with low and medium donation rates.
Recognising these limitations, the LEOPARD project aims to design and validate the following innovative solutions:
An AI-based allocation algorithm: Develop an AI-based LEOPARD predictive algorithm that outperforms current models in stratifying patients based on mortality risk. This algorithm will be proposed to OSOs to drive more efficient organ allocation.
DC & HCC LEOPARD calculators: Create calculators for both DC and HCC candidates, aiding professionals in making complex decisions regarding patient prioritisation for liver transplantation.
OMICS/radiomics integration: Integrate predictive signatures from OMICs and radiomics into a prototype 3rd-generation exploratory model exploring how OMICS may further enhance the accuracy of the predictive algorithm.
Beyond its technical innovations, the LEOPARD project aspires to contribute to a harmonised European approach to organ offering schemes, ultimately reducing mortality rates and establishing Europe as a leader in organ allocation schemes.
LEOPARD TEAM
Project Updates
Patient recruitment is now officially underway for the LEOPARD Training and Validation Data Collection Study (TVDCS). This study aims to gather a comprehensive dataset on liver transplantation candidates across Europe to support the development and validation of LEOPARD’s second-generation AI-based predictive models for mortality and dropout risk.
Recruitment follows a longitudinal, multicentre, prospective cohort study design, divided into two phases:
Training/Development Set: Data collection from 3,000 patients listed for transplantation across 50 centres in seven countries.
Validation Set: Data collection from an additional 1,500 patients listed in the same 50 centres.
This large-scale study will provide the critical data needed to refine and validate AI-driven predictive models, ultimately improving patient outcomes.
Project Timeline
click on the image to enlarge on laptop/desktop
Or download the PDF
LEOPARD Overview
Download the project overview for more information