[1715] A Triple Stain of Reticulin, Glypican-3, and Glutamine Synthetase Is Useful for the Diagnosis of Liver Lesions

Benjamin J Swanson, William L Marsh, Jr., Wendy L Frankel. Ohio State University Wexner Medical Center, Columbus, OH

Background: The histologic diagnosis of mass lesions of the liver can be difficult, especially when there is limited material in a biopsy. Here, we evaluated the usefulness of a triple stain of reticulin, glypican-3 (Gly3) and glutamine synthetase (GS) in whole tissue slides (WTS) and tissue microarrays (TMA) of hepatocellular carcinoma (HCC), hepatic adenoma (HA), non-neoplastic liver tissue adjacent to hepatocellular carcinoma (Liv), and focal nodular hyperplasia (FNH).
Design: WTS of liver masses were retrieved from the archives including HCC (n=16), HA (n=10) and FNH (n=13). In addition, TMA were constructed to mimic a small biopsy of HCC (n=19), HA (n=16), Liv (n=20) and FNH (n=13). A triple stain of reticulin followed sequentially by immunostains for Gly3 (alkaline phosphatase red chromogen) and GS (diaminobenzidine chromogen) was performed on all WTS and TMA. Cases were evaluated for reticulin (intact, lost), Gly3 (positive if >10% or negative) and GS (perivenular, map-like, diffuse if >50% or negative). Sensitivity and specificity in TMA cases were determined using known diagnoses from the resection specimen.
Results: Expression of reticulin, Gly3 and GS in both WTS and TMA cases is shown.

Reticulin, Gly3 and GS expression in Liver Lesions
WTS HA10/10 (100%)0010/10 (100%)7/10 (70%)01/10 (10%)
TMA HA15/16 (94%)1/16 (6%)016/16 (100%)8/16 (50%)01/16 (6%)
WTS FNH13/13 (100%)0013/13 (100%)013/13 (100%)0
TMA FNH13/13 (100%)0013/13 (100%)011/13 (85%)2/13 (15%)
WTS HCC1/16 (6%)15/16 (94%)8/16 (50%)8/16 (50%)1/16 (6%)09/16 (56%)
TMA HCC2/19 (10%)17/19 (90%)11/19 (58%)8/19 (42%)0015/19 (79%)
TMA Liv20/20 (100%)0020/20 (100%)15/20 (75%)00

The results in WTS and TMA are similar. However, map-like and perivenular GS positivity were decreased in TMA FNH and HA as compared to WTS, respectively, likely due to sampling error. Sensitivity and specificity in TMA HCC was compared to HA. Loss of reticulin was 94% sensitive and 94% specific for HCC. Gly3 positivity was 58% sensitive and 100% specific for HCC. Diffuse GS positivity was 79% sensitive and 94% specific for HCC. The sensitivity and specificity in TMA FNH was compared to HA. Intact reticulin was 100% sensitive and 6% specific for FNH. Gly3 negativity was 100% sensitive and 0% specific for FNH. A map-like GS positivity was 85% sensitive and 100% specific for FNH.
Conclusions: These results support the use of this triple stain in the differential diagnosis of HCC, HA and FNH in both WTS and biopsies.
Category: Liver

Tuesday, March 5, 2013 9:30 AM

Poster Session III # 195, Tuesday Morning


Close Window