To evaluate the combination of erlotinib and bevacizumab in subjects with hepatocellular carcinoma (HCC) who are not candidates for local therapy. Twenty-one subjects with metastatic or inoperable HCC who had not received local or systemic therapy were treated with 15 mg/kg bevacizumab every 3 weeks and a daily dose of 150 mg oral erlotinib. The primary endpoint was progression-free survival (PFS) at 27 weeks. The secondary endpoints were median time to progression and median overall survival. Twenty-one subjects were enrolled. Eighteen were evaluable for the primary endpoint; all subjects were evaluable for toxicity. The median age was 60 years (range, 33 to 81 y). Five subjects (28%) were progression free at 27 weeks (90% confidence interval (CI), 12%-50%). Median time to progression was 2.57 months (95% CI, 2.13-4.20 mo). Median overall survival was 8.33 months (95% CI, 5.73-13.97 mo). Two subjects withdrew consent, and 1 subject did not have adequate baseline scans. The 28% progression-free survival rate at 27 weeks was not significantly higher than the recent historical control rate of 20% observed on the placebo arm of the Sorafenib Hepatocellular Carcinoma Assessment Randomized Protocol trial (P=0.28). The combination of bevacizumab and erlotinib does not appear to have sufficient efficacy in patients with unresectable and metastatic HCC not amenable to local therapy, and may not warrant further investigation. However, this could be evaluated as an alternative to those intolerant to sorafenib therapy.