Scintigraphic liver function and transient elastography in the assessment of patients with resectable hepatocellular carcinoma.

BACKGROUND: Hepatobiliary scintigraphy (HBS) is used to quantify total and regional liver function. Transient elastography (TE) provides a non-invasive alternative to percutaneous biopsy to assess liver fibrosis and cirrhosis. This study aims to determine the correlation between HBS and histopathology of liver parenchyma, and to compare these with TE in patients with resectable hepatocellular carcinoma (HCC). METHODS: Patients who underwent surgery for HCC between 2000 and 2016 after preoperative HBS were included. Non-tumorous liver tissue was evaluated for inflammation, steatosis, ballooning, siderosis and fibrosis. Correlation analysis was performed between HBS results and histopathological scoring. These were also compared with TE and surgical outcomes. RESULTS: 71 patients underwent preoperative HBS of whom 24 also had TE. HBS correlated with portal and lobular inflammation as well as fibrosis. TE correlated with portal and lobular inflammation, ballooning and fibrosis. A significant correlation was found between HBS and TE. No association was found with overall postoperative morbidity and mortality. CONCLUSION: HBS and TE show a moderate to strong correlation. HBS and TE share discriminatory features of histopathological scoring and show a weak to moderate correlation with hepatic inflammation and fibrosis.

PD-L1, Galectin-9 and CD8+ tumor-infiltrating lymphocytes are associated with survival in hepatocellular carcinoma.

Novel systemic treatments for hepatocellular carcinoma (HCC) are strongly needed. Immunotherapy is a promising strategy that can induce specific antitumor immune responses. Understanding the mechanisms of immune resistance by HCC is crucial for development of suitable immunotherapeutics. We used immunohistochemistry on tissue-microarrays to examine the co-expression of the immune inhibiting molecules PD-L1, Galectin-9, HVEM and IDO, as well as tumor CD8+ lymphocyte infiltration in HCC, in two independent cohorts of patients. We found that at least some expression in tumor cells was seen in 97% of cases for HVEM, 83% for PD-L1, 79% for Gal-9 and 66% for IDO. In the discovery cohort (n = 94), we found that lack of, or low, tumor expression of PD-L1 (p < 0.001), Galectin-9 (p < 0.001) and HVEM (p < 0.001), and low CD8+TIL count (p = 0.016), were associated with poor HCC-specific survival. PD-L1, Galectin-9 and CD8+TIL count were predictive of HCC-specific survival independent of baseline clinicopathologic characteristics and the combination of these markers was a powerful predictor of HCC-specific survival (HR 0.29; p <0.001). These results were confirmed in the validation cohort (n = 60). We show that low expression levels of PD-L1 and Gal-9 in combination with low CD8+TIL count predict extremely poor HCC-specific survival and it requires a change in two of these parameters to significantly improve prognosis. In conclusion, intra-tumoral expression of these immune inhibiting molecules was observed in the majority of HCC patients. Low expression of PD-L1 and Galectin-9 and low CD8+TIL count are associated with poor HCC-specific survival. Combining immune biomarkers leads to superior predictors of HCC mortality.