Since our interim report of the intention-to-treat outcome of orthotopic liver transplantation (OLT) for hepatocellular carcinoma (HCC), we have performed a follow-up analysis of an expanded cohort of 70 patients to further assess whether the observed pattern and predictors of dropout are consistent with the rationale behind current HCC-adjusted Model for End Stage Liver Disease (MELD) organ allocation scheme. All except one patient had pretransplantation staging meeting our proposed expanded criteria—a single lesion ≤6.5 cm, or three or fewer lesions none >4.5 cm and total tumor diameter ≤8 cm. Thirty-eight patients received OLT. The cumulative probabilities of dropout at 6, 12, and 18 months were 7.2%, 37.8%, and 55.1%, respectively. The respective dropout probabilities would have been 11.0%, 57.4%, and 68.7% if the United Network for Organ Sharing (UNOS) criteria for exclusion (single lesion ≤5 cm or three or fewer lesions none >3 cm) were applied. Predictors of dropout with either criteria included three tumor nodules and a single lesion >3 cm at initial presentation, whereas preoperative chemoembolization or ablation therapies were associated with a lower risk for dropout only when applying the UNOS criteria for patient exclusion. In the subgroup with two or three lesions or a solitary tumor >3 cm, the cumulative probabilities of dropout were nine-fold higher than those with a single lesion ≤3cm (P = .004). In conclusion, the low dropout rate in the first 6 months and the differing dropout risks based on tumor characteristics support further refinements in the HCC-adjusted MELD organ allocation scheme. (Liver Transpl 2003;9:684-692.)