Liver transplantation for advanced hepatocellular carcinoma using poor tumor differentiation on biopsy as an exclusion criterion.

Background: Liberal acceptance criteria are used when offering liver transplantation (LTx) for treatment of hepatocellular carcinoma (HCC) at our center. This provides a unique opportunity to assess outcomes in a large North American series of patients with advanced tumors.
Objective: We hypothesized that acceptable survival rates can be achieved with LTx for any size or number of HCC provided that (a) imaging studies ruled out vascular invasion; (b) the HCC was confined to the liver; and (c) the HCC was not poorly differentiated on biopsy.
Methods: Survival, based on pretransplant imaging staging, was compared between 189 Milan Criteria (M) and 105 beyond Milan Criteria (M+) HCC patients who received an LTx between 1996 and 2008.
Results: Imaging understaged 30% of the M group and over staged 23% of the M+ group. There was no difference in the 5-year overall survival in the M (72%) and M+ (70%) groups or 5-year disease-free survival in the M (70%) and M+ (66%) groups. The introduction of a protocol for a biopsy to exclude patients with poorly differentiated tumors and use of aggressive bridging therapy improved overall survival in the M+ group (P = 0.034). Serum alpha-fetoprotein more than 400 at LTx was associated with poorer disease-free survival (hazard ratio: 2.3; P = 0.031).
Conclusions: Cross-sectional imaging did not reliably stage patients with HCC for LTx. A protocol using a biopsy to exclude poorly differentiated tumors and aggressive bridging therapy achieved excellent survival rates with LTx for otherwise incurable advanced HCC, irrespective of tumor size and number.