Retrospective clinicopathologic correlation of gross tumor size of hepatocellular carcinoma: implications for stereotactic body radiotherapy.
PURPOSE: To determine the degree of correlation between radiographic size and true gross pathologic size for subjects with primary hepatocellular carcinoma (HCC). METHODS AND MATERIALS: This analysis included 18 patients with 27 tumors who underwent either partial hepatectomy or orthotopic liver transplantation for HCC at the University of Colorado Hospital between 1997 and 2002. Preoperative imaging was performed using computed tomography (CT) or magnetic resonance imaging (MRI). After surgical resection the size of each tumor on gross pathologic examination was recorded. The maximal measurement in one dimension on axial imaging and pathologic examination was extracted for statistical analysis. The clinical and pathologic sizes were compared using a percent size difference (%Deltasize) as an end point for each patient. A regression analysis was applied to study the association between pathologic and clinical size. RESULTS: The median radiographic size was 2.90 cm (range 1.2-4.9). The median pathologic size was 2.50 cm (range 1-4.8). The radiographic size was larger than or equal to the pathologic size in 22/27 tumors (81%) and smaller in 5/27 (19%) tumors. The median %Deltasize was 17.5% (range -20-144%). Overall, the radiographic and pathologic sizes were positively correlated (r = 0.8). This correlation was not affected by choice of imaging modality (CT versus MRI, P = 0.71) or time of preoperative imaging (0-4 weeks versus 4-8 weeks before surgery, P = 0.61). CONCLUSIONS: Our study shows that in most instances (81%), imaging by CT or MRI overestimates true gross pathologic size of HCC. Nineteen percent of tumors appeared smaller on preoperative imaging than on the final pathologic specimen. Radiation therapy utilizing a 0.5 or 1.0 cm margin around the radiographic tumor would have encompassed the gross pathologic tumor in 93% and 100% of cases, respectively.