Characterization of Hepatocellular Carcinoma Patients with FGF19 Amplification Assessed by Fluorescence in situ Hybridization: A Large Cohort Study

Abstract

Background: FGF19 amplification is a relatively novel type of genetic aberration that has been proposed to be a driver of hepatocarcinogenesis. Selective inhibitors of FGFR4, a receptor of FGF19, have been developed as targeted therapies for hepatocellular carcinoma (HCC). Despite the role of FGF19 in mediating HCC progression, the clinicopathological characterization of patients exhibiting FGF19 amplification remains unclear. Immunohistochemical staining is the simplest and most widely used method of identifying aberrations in the FGF19 gene, although its specificity is very low.

Methods: This study investigated the prognostic significance of FGF19 amplification in a large cohort of 989 HCC patients using fluorescence in situ hybridization (FISH), which has a high degree of specificity. In addition, FISH data from formalin-fixed, paraffin-embedded sections were compared with copy number variation (CNV) data obtained from fresh frozen sections to validate the use of FISH as a diagnostic tool.

Results: FGF19 amplifications were detected by FISH in 51 (5.15%) of the 989 patients, and were independently associated with poor survival and a higher risk of tumor recurrence, as well as with poor prognostic factors such as a high α-fetoprotein level, hepatitis B or C virus infection, a large tumor size, microvascular invasion, and necrosis. In addition, FGF19 amplification was associated with TP53 mutation, and was mutually exclusive with CTNNB1 mutation. The results of the FISH and CNV analyses exhibited a significant concordance rate of 96% (κ = 0.618, p < 0.001).

Conclusions: These data indicate that FGF19 amplification represents a unique molecular subtype associated with poor prognostic characteristics, which supports the hypothesis that the FGF19-FGFR4 signaling pathway plays an important role in hepatocarcinogenesis. We have also demonstrated that FISH is a viable alternative to CNV analysis, offering a number of advantages in the clinical setting.

Keywords: FGF19; Fluorescence in situ hybridization; Hepatocellular carcinoma; High copy number amplification; Prognosis.

Fig. 1.

Fluorescence in situ hybridization micrographs showing FGF19 (red signal) and CEP11 (green signal) fluorescence in tumor cells and non-tumor cells. aFGF19 positive in tumor cells: an increase in the number of red signals in tumor cells, indicating amplification of FGF19 (left upper inset). Normal FGF19 or CEP11 signals served as the internal positive control in lymphocytes (left lower inset). Magnification, ×1,000. b, cFGF19 negative in tumor cells (b) and FGF19 negative in non-tumor cells (c): no increase in red signals in tumor and non-tumor cells, indicating no amplification of FGF19, respectively (insets). Magnification, ×1,000.

Fig. 2.

Recurrent gene mutations in patients with hepatocellular carcinoma exhibiting/lacking FGF19 amplification (as determined by fluorescence in situ hybridization analysis).

Fig. 3.

Comparison of survival rates according to FGF19 amplification status. The positive group, with FGF19 amplification, was associated with poorer overall survival (a) and disease-free survival (b) rates.

Fig. 4.

Comparison of the detailed results between fluorescence in situ hybridization (FISH) and copy number variation (CNV) analyses. Among 227 cases, 219 (96%) were matched by the CNV analysis and FISH results. Of the 8 (3%) discrepant cases, 3 (1%) were positive by FISH with one copy gain in the CNV analysis, whereas 5 (2%) were negative by FISH with high copy amplification in the CNV analysis.