Assessing Competing Risks for Death Following Liver Transplantation for Hepatocellular Carcinoma

Abstract

Hepatocellular carcinoma (HCC) as an indication to liver transplant (LT) started as palliative treatment, then moved to potentially curative anti-cancer therapy and more recently entered the era of competition with non-cancer indications, consequent to the need of the society to target equal distribution of the limited resource of donated organs among different indications. Nowadays HCC is a leading indication to LT, currently representing up to 50% of the indications in most transplant Centers. The risk of post-transplant death and the causes of mortality significantly vary along the post-transplant follow-up. Overall, the main causes of death after LT are multiple organ failure and cerebrovascular, cardiovascular, pulmonary, and renal complications. However, after the first post-LT year, mortality for technical complications, infections and general complications significantly decrease, while recurrence of primary liver diseases (particularly malignancies) increase, turning to be the main causes of death. In studies with time-to-event or survival outcomes, a competing risk is an event whose occurrence precludes the occurrence of the primary event of interest. In the setting of LT for HCC, when the primary outcome of interest is death due to HCC recurrence, death due to causes different from this serves as a competing event because subjects who die from such different causes are no longer at risk of death due to HCC recurrence. The introduction of HCC-specific survival as a primary endpoint in studies assessing the outcomes of LT for HCC allows the identification of independent oncologic determinants of post-LT survival and their relative weight on patients’ prognosis. In this view, a continuous model based on level of AFP, tumor size and tumor number that allows to determine the risk of death from HCC-related factors after liver transplantation (www.hcc-olt-metroticket.org/) has been recently developed. Since the endpoint of HCC-specific survival is not influenced by the changes observed in short-term post-LT survival (thanks to advances in the clinical management) nor in long-term post-LT survival (thanks to the introduction of effective treatments achieving control of hepatitis B and C viruses), a model predicting HCC-specific survival will be an helpful prognostic tool in the context of the changing scenarios of LT for HCC.