BACKGROUND: Chronic liver disease is a major cause of premature death in sub-Saharan Africa. Efficacy of antiviral therapy among patients with hepatitis B virus (HBV)-related cirrhosis is not well established in Africa. We described the clinical characteristics and outcomes of patients with cirrhosis and hepatocellular carcinoma in The Gambia and assessed the impact of tenofovir disoproxil fumarate (TDF) on survival of HBV-infected patients with cirrhosis. METHODS: In this prospective cohort study, we followed up adults who were consecutively diagnosed with cirrhosis or hepatocellular carcinoma between 2012 and 2015 in The Gambia, west Africa. Patients with chronic HBV infection and cirrhosis, without hepatocellular carcinoma, were offered TDF. Primary outcome was overall survival. To determine the effect of TDF on survival, we performed a Cox proportional hazard regression model with inverse probability of treatment weighting (IPTW) based on propensity score. FINDINGS: Of 529 patients enrolled in this study, 336 patients (252 with hepatocellular carcinoma and 84 with cirrhosis) were analysed. Patients were predominantly male (253 [75%] men and 83 [25%] women), with a median age of 42 years (IQR 33-55). 276 (84%) of 327 of patients with data were positive for HBV biomarkers, 31 (10%) of 311 were positive for hepatitis C virus antibodies, and 22 (10%) of 223 were positive for hepatitis D virus antibodies. 64% of patients with hepatocellular carcinoma had multifocal tumour, with a median size of 7·5 cm (IQR 5·4-10·8). 173 patients with hepatocellular carcinoma and 70 patients with cirrhosis were included in the survival analysis. Median survival was 1·5 months (95% CI 1·1-2·0) in patients with hepatocellular carcinoma and 17·1 months (11·2-24·0) in patients with cirrhosis (log-rank p<0·0001). In patients with hepatocellular carcinoma, ascites (hazard ratio [HR] 1·78, 95% CI 1·21-2·60), partial or complete portal thrombosis (HR 2·61, 1·58-4·30), and platelet count (HR 1·80, 1·19-2·70) were independent predictive factors of mortality at baseline. In HBV-infected patients with cirrhosis, median turnaround time between cirrhosis diagnosis and TDF initiation was 4·9 months (IQR 3·2-7·3). In IPTW analysis, TDF treatment was associated with improved survival in patients with HBV-related cirrhosis (adjusted HR 0·14, 0·06-0·34; p<0·0001). INTERPRETATION: These results highlight poor survival of patients with cirrhosis or hepatocellular carcinoma as well as the effectiveness of TDF in reducing the premature mortality of patients with cirrhosis and HBV infection. Interventions for early diagnosis and treatment of cirrhosis as well as screening programmes for hepatocellular carcinoma are urgently required in Africa. FUNDING: European Commission and Medical Research Council UK. TRANSLATION: For the French translation of the abstract see Supplementary Materials section.