Slow-growing tumors have the best prognosis. It is increasingly clear that hepatitis B and C play a role in the develop-Hepatocellular carcinoma (HCC), one of the most frequent ment and progression of HCC both before and after liver malignant tumors in the world, can be fatal if untreated. transplantation. 13-16 The influence of immunosuppression on Only surgical removal of the tumor offers a chance of longde novo and recurrent posttransplant malignancies, once a term survival or a cure. Two therapeutic options are availsubject of speculation, has been confirmed. 17 Understanding able: partial liver resection or total hepatectomy with liver the influence of these various factors on malignant progrestransplantation, including various technical modifications in sion requires basic research. Second, an accurate assessment between or beyond, e.g., ex vivo bench surgery and multivisof prognosis requires imaging techniques to detect all intraceral resection, respectively. 1-4 The chosen approach depends and extra-hepatic cancer and to accurately determine the mainly on the technical resectability of the tumor and on the tumor stage preoperatively. Vascular invasion by tumor, functional status of the residual liver.
which is a highly significant prognostic factor, can only be A critical prerequisite for comparing results among studies determined by a thorough, histological examination. From is an accurate assessment of clearly defined clinical and repeated and proven clinical experience, no existing radiopathological factors and a common system of tumor classifilogical method provides this information and even surgical cation. The need for this cannot be over-emphasized. The exploration is less than definitive. 18-20 The lingering possibil-TNM system, already proposed, sets forth exact criteria for ity of microscopic invasion is one of the arguments for adjuassessing the intrahepatic extent of tumor (size, number, vant chemo-or radiotherapy, although the present regimens location, and vascular invasion) as well as the extrahepatic are marginally beneficial. 21-23 spread to regional lymph nodes and distant sites. 5 The im-This is the background of the paper by Marsh et al., pubpairment of liver function caused by underlying cirrhosis of lished in this issue of HEPATOLOGY. 24 In that study, the viral, alcoholic, or other origin is also important with regard group's major goal was to fill an important gap in the process to early morbidity, early mortality, and long-term survival.
of decision-making and pretransplant recipient selection; Uni-and multivariate analyses of large series have yielded they had a large patient group with which to work. A total a number of factors related to tumor recurrence and patient of 178 recipients, including 63 cases with co-existing liver survival. 6-12 While these studies are mostly retrospective, sevdiseases, were eligible for this pilot study. After a followeral addressed the role of liver transplantation in HCC. Once up of 0.12 to 14.2 years (median, 3.35 years), 71 (40%) thought to be the ideal therapy for HCC, transplantation was experienced tumor recurrence, and for virtually all of these eventually contraindicated because of a high rate of tumor (70 of 71 patients) the time interval was less than the critical recurrence. More recently, there is new clinical and scientific period of 3 years. An analysis of the data yielded the following interest in this procedure for HCC which is based primarily five clinicopathological risk factors: gender, tumor number, on better patient selection from new imaging techniques and location, size, and vascular infiltration, all of which are in on increased experience with transplantation combined with agreement with the TNM classification and had been identineo-and adjuvant chemotherapy. To date, however, it seems fied in previous studies. Based on the assumption that the easier to identify patients who are at a high risk of failure standard proportional hazards models, such as the Kaplanthan to predict those at low risk. Despite very sophisticated Meier analysis, do not consider changes of covariates over statistical analyses of hundreds, or even thousands, of patime, artificial neural networks were constructed for analyztients, the predictive value of current parameters is suboptiing the risk of recurrence separately within the first, second, mal for both the low-and high-risk groups. Far-advanced and third year after transplantation. According to these modcancer cases (TNM stage IVA) can be resected for survival; els the selected risk factors allow for the perfect stratification conversely, stage-II patients with small solitary lesions may of patients into the three following groups: 1) those at low experience early relapse and death. risk of recurrence; 2) those at high risk of recurrence; and Thus, the real dilemma in liver transplantation for HCC is 3) those whose risk was indeterminant. Marsh et al. also to predict the postoperative course in the individual patient.
attempt to assess the influence of postoperative chemother-Insight into this pressing question is very difficult, if not apy on survival and tumor recurrence. Of the 51 patients impossible and caused by various imponderables, two of who received various protocols, 42 had intra-arterial or intrawhich are as follows. First, the differing biological behavior venous adjuvant treatment. Overall, there was no significant effect of this treatment. When the data were broken down by risk group, it was found that chemotherapy did not benefit Abbreviation: HCC, hepatocellular carcinoma.