TUMOR RELATED SURVIVAL AFTER LIVER TRANSPLANTANTION OR RESECTION FOR HCC: A COMPETING RISK ANALYSIS WITH AN INTENTION TO TREAT ANALYSIS PERSPECTIVE

Background: Many patients with Hepatocellular Carcinoma (HCC) present at diagnosis a tumor stage suitable for both Liver Resection (LR) and Liver Transplantation (LT). LR is generally safer, available for all, schedulable at the best timing, but less radical and not able to cure the underlying cirrhosis. Contrarily, LT is much more invasive, unavailable for all the candidates, performed at unpredictable time, but with more extended radicality and able also to cure the underlying cirrhosis. Until now, what is the best candidate for LR or LT upfront is extremely unclear. The aim of the study was to validate a model to predict tumor results after LR or LT for patients eligible for both. Methods: All the 2640 consecutive cases of LR or LT for HCC at the 4 centers involved into the study were collected. After a propensity score matching (PSM), an homogeneous patient’s cohort was selected and compared between LR and LT groups. An intention-to-treat (ITT) analysis with a competing risk model was applied comparing the tumor related deaths of patient groups. Results: Between 2005 and 2015, 551 LRs and 580 LT candidates (LTc) were selected for the study. Of the 580 LTc, 512 were effectively transplanted, whereas 68 (11.7%) were dropped due to tumor progression. After a PSM, 101 high risk patients were, respectively, selected for each groups: LR and LT. Variables considered for the risk assignment were: alfafetoprotein, tumor size, tumor number, AST, MELD, and age at tumor diagnosis. The 3- and 5-years cumulative incidence of tumor related death was, respectively, 23% and 31% vs 5% and 9% for LR and LT group (Pvalue = .002). Conclusions: The high risk patients’ cohort has significantly better ITT tumor-related survival with a LT rather than a LR upfront. The microvascular invasion or HCC satellitosis may significantly increase the risk of tumor related death even in low and intermediate risk patients who have received LR. The ab-initio LT should be considered in these reconverted risk patients