From large to small: the immunohistochemical panel in the diagnosis of early hepatocellular carcinoma.

AIMS: The aims of this study were to: validate the use of the immunohistochemical (IHC) markers glutamine synthetase (GS), glypican-3 (GPC3), heat shock protein-70 (HSP70) and enhancer of zeste homologue 2 (EZH2) in liver biopsies for the differential diagnosis between small hepatocellular carcinoma (HCC) and non-neoplastic liver nodules, with special attention to <10-mm nodules; and assess the actual sensitivity and specificity of the single markers, and their combination, in needle biopsies.

METHODS AND RESULTS: One hundred liver nodules, i.e. 66 HCCs and 34 non-neoplastic nodules, were prospectively collected from 43 consecutive orthotopic liver transplantation patients, and subjected to 'backtable' needle biopsies directly on surgical specimens. IHC evaluation was semi-automatically performed with a Benchmark Ultra immunostainer. The morphological and IHC diagnosis in surgical specimens was considered to be the gold standard. GS, GPC3, HSP70 and EZH2 showed 16.6%, 10.7%, 28.8% and 62.1% decreases in sensitivity, respectively, from surgical specimen to needle biopsy. Higher decreases were observed in <10-mm nodules. In 18 HCCs with no morphological diagnostic features of malignancy in biopsies, GPC3 or GS were positive in 16; in seven HCCs, neither morphology nor IHC evaluation ruled out the differential diagnosis made on the basis of needle biopsy.

CONCLUSIONS: We present for the first time a direct comparison between surgical specimens and needle biopsies to confirm the usefulness and reproducibility of the most widely used antibodies for the diagnosis of small liver nodules. Our results support the use of IHC evaluation in biopsies for the diagnosis of small liver lesions, although the IHC panel could also give negative results in the presence of obvious HCC, and the possibility of false positives should always be considered.