NATURAL-HISTORY OF SMALL UNTREATED HEPATOCELLULAR-CARCINOMA IN CIRRHOSIS - A MULTIVARIATE-ANALYSIS OF PROGNOSTIC FACTORS OF TUMOR-GROWTH RATE AND PATIENT SURVIVAL

We analyzed the growth pattern of tumor masses and the survival of 39 asymptomatic Italian patients with a total of 59 small (less-than-or-equal-to 5 cm in diameter) hepatocellular carcinomas arising from cirrhosis. The total length of the observation period ranged from 90 to 962 days, with an average of 364 +/- 229 (mean +/- S.D.). Doubling time ranged from 27.2 to 605.6 days (mean +/- S.D., 204.2 +/- 135; median = 171.6 days). Three different growth patterns were recognized: (a) tumors with no or very slow initial growth pattern (doubling time > 200 days), 10 cases (37%); (b) tumors with declining growth rate over time, 9 cases (33.4%); and (c) tumors with almost constant growth rate, 8 cases (29.6%). Using the stepwise discriminant analysis, we found a score based on albumin, alcohol intake, number of nodules, echo pattern and histological type that allowed a correct prediction of short doubling time (less-than-or-equal-to 150 days) in 55.6%, medium doubling time (151 to 300 days) in 60% and long doubling time (> 300 days) in 100% of cases. The estimated survival rate of the 39 patients, calculated by the Kaplan-Meier method was 81% at 1 yr, 55.7% at 2 yr and 21% at 3 yr. Stepwise discriminant analysis showed that a score based on sex, HBsAg status, alcohol consumption, ascites, gamma-glutamyltranspeptidase, prothrombin time, Child-Pugh class and all the sonographical parameters could predict 2-yr survival in 100% of cases. We conclude that great variability of growth patterns exists among and within small hepatocellular carcinomas. Prediction of subsequent growth rate is unreliable in most cases. The sonographical characteristics, together with the histological features, can, however, help in identifying cases with long doubling time (> 300 days). The discriminant analysis on survival of cirrhotic patients with small hepatocellular carcinomas demonstrates that the underlying liver disease plays a key role in the long-term survival probability.