Prediction of Protumorigenic Effects after Image-Guided Radiofrequency Ablation of Hepatocellular Carcinoma Using Biomarkers

Purpose: To perform radiofrequency (RF) ablation of hepatocellular carcinoma (HCC) and to assess serological and histopathological markers of tumorigenesis in distant untreated tumors to determine whether these were associated with unfavorable outcomes such as early relapse and increased biological aggressiveness. Materials and Methods: The study cohort comprised 13 patients from a prospective single-arm study. All patients underwent 2 ablation sessions of multifocal HCC nodules 14 days apart. Core biopsy samples of untreated tumors were acquired at baseline and at the time of the second ablation session. Samples were stained immunohistochemically with Ki-67 (proliferation) and CD34 (microvasculature). Blood plasma was obtained at baseline and 2 days after the initial ablation session and analyzed for hepatocyte growth factor (HGF), vascular endothelial growth factor C, and angiopoietin-2 using an enzyme-linked immunosorbent assay. The clinical follow-up period ranged from 7 to 25 months. Patients were stratified as responders (complete remission or limited and delayed recurrence at >6 months; n = 6) or nonresponders (any recurrence within 6 months or >3 new tumors or any new tumor of >3 cm thereafter; n = 7). Results: In 3 of 7 nonresponders, the Ki-67 index markedly increased in untreated tumors, whereas Ki-67 was stable in all responders. Microvascular density strongly increased in a single nonresponder only. HGF and angiopoietin-2 increased by >30% in 3 of 7 and 4 of 7 nonresponders, respectively, whereas they were stable or decreased in responders. Overall, ≥2 biomarkers were elevated in 6 of 7 (85.7%) nonresponders, whereas 4 of 6 responders demonstrated no increased biomarker and 2 patients demonstrated increase in 1 biomarker only (P =.002). Conclusions: RF ablation of HCC can produce protumorigenic factors that induce effects in distant untreated tumors. These may potentially function as biomarkers of clinical outcome.