Prognostic factors in patients with hepacellular carcinoma and macrovascular invasion

LOPEZ P; ROAYAIE S; UVA P; PATEL P; SCHWATZ J; SHAO T; DEL POZO A; SCHWARTZ M; LLOVET J

Boston

Congreso; Annual Meeting of the AASLD; 2007

American Association for the Study of Liver Diseases

Purpose: To analyze the outcome and prognostic factors of a cohort of patients with hepatocellular carcinoma (HCC) and macrovascular invasion (MVI) and establish subgroup categories with different outcomes. Methods: This is a single center retrospective study from March 2001 to December 2005 of patients with newly diagnosed HCC with MVI. The study included patients with HCC and MVI diagnosed by imaging studies (dual phase CT scan or MRI) and reliable clinical variables available. HCC was diagnosed according to the consensus statements established by the EASL and AASLD. Predictors of survival were identified by Kaplan Meyer analysis and Mantel-Cox comparisons. Multivariate Cox proportional hazard analysis was conducted with variables showing significant association with survival at univariate analysis. Results: During the study period, 231 patients out of the 936 HCC patients diagnosed in our center presented with MVI (24.6%). Among them, 199 pts met the inclusion criteria:. Seventy were Child Pugh-A (35%), 86 Class B (43%) and 42 Class C (21%). 114 (58%) were ECOG 0 (asymptomatic) and 84 (42%) were symptomatic. HCC was solitary in 110 patients (55.3%). Main portal trunk was involved in 110 (51%), first portal branch in 79 (40%) and a  second branch in 18 (9%). Hepatic vein invasion occurred in 39 (20%). Extra hepatic spread was present in 47 patients(23.6%). 143 patients were Barcelona Clinic Liver Cancer Program (BCLC) stage C and 56 BCLC stage D. 115 (57.8%) received treatment after MVI diagnosis, including resection in 34 cases. Median survival for the entire cohort was 4.6 months, the 1 and 2 years survival rate was of 17% and 5%, respectively. Median survival of BCLC C was 6.2 months and BCLC D was of 2.2 months. In BCLC C patients, ECOG 0 (p<0.01), absence of extra hepatic spread (p<0.01) and resection (p=0.01) were independent prognostic factors. Two subgroup of patients with good [asymptomatic patients without extra hepatic spread; n=79) and poor prognosis [n=64] were established. Median survival of these groups was of 8.1 months vs 4.5 months, respectively (p<0.001). Median survival of the 34 patients resected in the good prognosis group was of 10.2 months. Conclusion: HCC patients with macrovascular invasion have a heterogenous outcome. Independent prognostic factors (ECOG and extrahepatic spread) enabled us to recognize two outcome subgroups, and provide the rationale for  stratification of patients included in clinical trials. Whether resection might benefit a subgroup of patients with MVI and well-preserved liver function needs to be elucidated in additional studies.