Alessandro Vitale1, Michael L Volk2, Tullia Maria De Feo3, Patrizia Burra4, Anna Chiara Frigo1, Rafael Ramirez Morales5, Luciano De Carlis6, Luca Belli7, Michele Colledan8, Stefano Fagiuoli8, Giorgio Rossi9, Enzo Andorno10, Umberto Baccarani11, Enrico Regalia12, Marco Vivarelli13, Matteo Donataccio14, Umberto Cillo1. 1. University Hospital of Padua, Italy. 2. Division of Gastroenterology and Hepatology, University of Michigan, Ann Arbor, USA. 3. Organs and Tissue Transplant Immunology Unit, Fond. IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy. 4. University Hospital of Padua, Italy. Electronic address: burra@unipd.it. 5. Veneto Institute of Oncology IOV - IRCCS, Padua, Italy. 6. Surgery and Transplantation, Ospedale Niguarda Ca' Granda, Milan, Italy. 7. Hepatology and Gastroenterology, Ospedale Niguarda Ca' Granda, Milan, Italy. 8. Gastroenterology and Transplantation Hepatology, Ospedali Riuniti, Bergamo, Italy. 9. Liver Transplantation Unit, Fond. IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy. 10. IRCCS San Martino, Genoa, Italy. 11. Medical Liver Transplant Unit, University of Udine, Italy. 12. Liver Transplantation, IRCCS INT, Milan, Italy. 13. Liver Transplantation Unit, Ancona Hospital, Italy. 14. Liver Transplantation Unit, Verona Hospital, Italy.
Abstract
BACKGROUND & AIMS: The current organ allocation system for liver transplantation (LT) creates an imbalance between patients with and without hepatocellular carcinoma (HCC). We describe a model designed to re-establish allocation equity among patient groups using transplant benefit as the common endpoint. METHODS: We enrolled consecutive adult patients entering the waiting list (WL group, n=2697) and undergoing LT (LT group, n=1702) during the period 2004-2009 in the North Italy Transplant program area. Independent multivariable regressions (WL and LT models) were created for patients without HCC and for those with stage T2 HCC. Monte Carlo simulation was used to create distributions of transplant benefit, and covariates such as Model for End-stage Liver Disease (MELD) and alpha-fetoprotein (AFP) were combined in regression equations. These equations were then calibrated to create an "MELD equivalent" which matches HCC patients to non-HCC patients having the same numerical MELD score. RESULTS: Median 5 year transplant benefit was 15.12 months (8.75-25.35) for the non-HCC patients, and 28.18 months (15.11-36.38) for the T2-HCC patients (p<0.001). Independent predictors of transplant benefit were MELD score (estimate=0.89, p<0.001) among non-HCC patients, and MELD (estimate=1.14, p<0.001) and logAFP (estimate=-0.46, p<0.001) among HCC patients. The equation "HCC-MELD"=1.27∗MELD - 0.51∗logAFP+4.59 calculates a numerical score for HCC patients, whereby their transplant benefit is equal to that of non-HCC patients with the same numerical value for MELD. CONCLUSIONS: We describe a method for calibrating HCC and non-HCC patients according to survival benefit, and propose that this method has the potential, if externally validated, to restore equity to the organ allocation system.
BACKGROUND & AIMS: The current organ allocation system for liver transplantation (LT) creates an imbalance between patients with and without hepatocellular carcinoma (HCC). We describe a model designed to re-establish allocation equity among patient groups using transplant benefit as the common endpoint. METHODS: We enrolled consecutive adult patients entering the waiting list (WL group, n=2697) and undergoing LT (LT group, n=1702) during the period 2004-2009 in the North Italy Transplant program area. Independent multivariable regressions (WL and LT models) were created for patients without HCC and for those with stage T2 HCC. Monte Carlo simulation was used to create distributions of transplant benefit, and covariates such as Model for End-stage Liver Disease (MELD) and alpha-fetoprotein (AFP) were combined in regression equations. These equations were then calibrated to create an "MELD equivalent" which matches HCC patients to non-HCC patients having the same numerical MELD score. RESULTS: Median 5 year transplant benefit was 15.12 months (8.75-25.35) for the non-HCC patients, and 28.18 months (15.11-36.38) for the T2-HCC patients (p<0.001). Independent predictors of transplant benefit were MELD score (estimate=0.89, p<0.001) among non-HCC patients, and MELD (estimate=1.14, p<0.001) and logAFP (estimate=-0.46, p<0.001) among HCC patients. The equation "HCC-MELD"=1.27∗MELD - 0.51∗logAFP+4.59 calculates a numerical score for HCC patients, whereby their transplant benefit is equal to that of non-HCC patients with the same numerical value for MELD. CONCLUSIONS: We describe a method for calibrating HCC and non-HCC patients according to survival benefit, and propose that this method has the potential, if externally validated, to restore equity to the organ allocation system.