Tamoxifen is not effective in good prognosis patients with hepatocellular carcinoma

Background: Large randomised clinical trials and systematic reviews substantiate that tamoxifen is ineffective in improving survival of patients with hepatocellular carcinoma (HCC). However, a recent report suggested that the drug might prolong survival among patients with well preserved liver function. The aim of this paper is to validate this hypothesis. Methods: We used the updated database of the phase 3 randomised CLIP-I trial that compared tamoxifen with supportive therapy. Primary endpoint was overall survival. Treatment arms were compared within strata defined according to the Okuda stage and the CLIP-score. Survival differences were tested by the Log-rank test. Results: Tamoxifen was not effective in prolonging survival in Okuda I-II subgroup (p = 0.501). Median survival times were equal to 16.8 (95% CI 12.7-18.5) months for tamoxifen and 16.8 (95% CI 13.5-22.4) months for the control arms; 1-year survival probabilities were equal to 58.8% (95% CI 51.7-65.8) and 59.4 (95% CI 52.5-66.2), respectively. Similar results were observed in the better CLIP subgroup (score 0/1), without evidence of difference between the two treatment arms (p = 0.734). Median survival times were equal to 29.2 (95% CI 20.1-36.4) months with tamoxifen and 29.0 (95% CI 23.3-35.2) months without; 1-year survival probabilities were equal to 80.9% (95% CI 72.5-89.3) with tamoxifen and 77.1% (95% CI 68.6-85.7) for the control arm. Conclusion: The recent suggestion that tamoxifen might be effective in the subgroup of patients with better prognosis is not supported by a reanalysis of the CLIP-I trial. Tamoxifen should no longer be considered for the treatment of HCC patients and future trials of medical treatment should concentrate on different drugs.