Systematic review of randomized trials for unresectable hepatocellular carcinoma: Chemoembolization improves survival

There is no standard treatment for patients with unresectable hepatocellular carcinoma (HCC). Survival benefits derived from medical interventions are controversial. The aim of this systematic review was to assess the evidence of the impact of medical treatments on survival. Randomized controlled trials (RCTs) that were published as full papers assessing survival for primary treatments of HCC were included. MEDLINE, the Cochrane Library, CANCERLIT, and a manual search from 1978 to May 2002 were used. The primary end point was survival, and the secondary end point was response to treatment. Estimates of effect were calculated according to the random effects model. Sensitivity analysis included methodological quality. We identified 61 randomized trials, but only 14 met the criteria to perform a meta-analysis assessing arterial embolization (7 trials, 545 patients) or tamoxifen (7 trials, 898 patients). Arterial embolization improved 2-year survival compared with control (odds ratio [OR], 0.53; 95% confidence interval [CI], 0.32-0.89; P =.017). Sensitivity analysis showed a significant benefit of chemoembolization with cisplatin or doxorubicin (OR, 0.42; 95% CI, 0.20-0.88) but none with embolization alone (OR, 0.59; 95% CI, 0.29-1.20). Overall, treatment induced objective responses in 35% of patients (range, 16%-61%). Tamoxifen showed no antitumoral effect and no survival benefits (OR, 0.64; 95% CI, 0.36-1.13; P =.13), and only low-quality scale trials suggested 1-year improvement in survival. In conclusion, chemoembolization improves survival of patients with unresectable HCC and may become the standard treatment. Treatment with tamoxifen does not modify the survival of patients with advanced disease.