Hepatocellular carcinoma (primary liver cancer) is a common cause of death from cancer world-wide. It is usually fatal in patients who cannot be treated with surgery or other local treatments. Tamoxifen is an anti-oestrogen drug, which has been tested in advanced hepatocellular carcinoma. The reviewers identified 10 trials assessing the effect of tamoxifen on survival, quality of life, tumour size, and treatment side effects in advanced hepatocellular carcinoma. Tamoxifen had no significant effect on survival or tumour size. Tamoxifen did not improve quality of life.
These data do not support the use of tamoxifen for patients with hepatocellular carcinoma. Further research on the effects of tamoxifen in hepatocellular carcinoma does not seem warranted.
Hepatocellular carcinoma (primary liver cancer) is the third commonest cause of cancer mortality world-wide. Survival is poor for patients with advanced disease. Trials of tamoxifen for hepatocellular carcinoma have conflicting results.
To conduct a systematic review of the literature to assess the effect of tamoxifen on overall survival, quality-of-life, tumour response, and treatment toxicity in people with advanced hepatocellular carcinoma.
We identified trials from The Cochrane Hepato-Biliary Group Controlled Trials Register (January 2004), The Cochrane Central Register of Controlled Trials on The Cochrane Library (Issue 3, 2003), and MEDLINE database (1966 to November 2003). We searched bibliographies of review articles and identified trials, and hand-searched abstracts from relevant other meetings.
All randomised clinical trials of treatment with tamoxifen compared to a control treatment without tamoxifen in people with hepatocellular carcinoma, including trials of tamoxifen versus placebo, tamoxifen versus best supportive care, and tamoxifen plus other treatment versus the same other treatment alone.
Three independent reviewers selected studies for inclusion, rated them for methodologic quality components (generation of allocation sequence; allocation concealment; blinding; and follow-up), and extracted data on the specified outcomes. Hazard ratios were derived for overall survival where possible. Meta-analysis was performed using a fixed-effect model.
Ten randomised trials randomising 1709 patients were included. Tamoxifen versus placebo/no intervention had no significant effect on overall survival (hazard ratio 1.05; 95% CI 0.94 to 1.16; P = 0.4). This comparison showed no statistical heterogeneity (P = 0.2 and I2 = 25.9%). Subgroup analysis showed that tamoxifen tended to increase mortality in trials with three adequate/three methodological components (hazard ratio 1.15; 95% CI 0.99 to 1.34; P = 0.06), showed no significant effect in trials with two adequate/three methodological components (hazard ratio 1.00; 95% CI 0.84 to 1.18; P = 0.98), and tended to reduce mortality in trials with one or less adequate/three methodological components (hazard ratio 0.82; 95% CI 0.60 to 1.12; P = 0.2), although this may have been confounded by the use of higher doses of tamoxifen in the better quality trials. Tamoxifen was associated with adverse effects. One trial measured patient quality of life, but the results were not reported in detail.