BACKGROUND In patients with hepatocellular carcinoma (hepatoma), the rate of recurrent and second primary hepatomas is high despite surgical resection and percutaneous ethanol-injection therapy. We developed an acyclic retinoid, polyprenoic acid, that inhibits hepatocarcinogenesis in the laboratory and induces differentiation and apoptosis in cell lines derived from human hepatoma. In a randomized, controlled study, we tested whether the compound reduced the incidence of recurrent and second primary hepatomas after curative treatment. METHODS We prospectively studied 89 patients who were free of disease after surgical resection of a primary hepatoma or the percutaneous injection of ethanol. We randomly assigned the patients to receive either polyprenoic acid (600 mg daily) or placebo for 12 months. We studied the remnant liver by ultrasonography every three months after randomization. The primary end point of the study was the appearance of a histologically confirmed recurrent or new hepatoma. RESULTS Treatment with polyprenoic acid significantly reduced the incidence of recurrent or new hepatomas. After a median follow-up of 38 months, 12 patients in the polyprenoic acid group (27 percent) had recurrent or new hepatomas as compared with 22 patients in the placebo group (49 percent, P = 0.04). The most striking difference was in the groups that had second primary hepatomas--7 in the group receiving polyprenoic acid as compared with 20 in the placebo group (P = 0.04 by the log-rank test). Cox proportional-hazards analysis demonstrated that as an independent factor, polyprenoic acid reduced the occurrence of second primary hepatomas (adjusted relative risk, 0.31; 95 percent confidence interval, 0.12 to 0.78). CONCLUSIONS Oral polyprenoic acid prevents second primary hepatomas after surgical resection of the original tumor or the percutaneous injection of ethanol.