Searchable abstracts of presentations at key conferences in endocrinology
Endocrine Abstracts (2004) 7 P76

BES2004 Poster Presentations Endocrine tumours and neoplasia (53 abstracts)

Galectin-3 staining of benign and malignant thyroid lesions - is it a useful diagnostic tool?

R Davies 1 , M Barakat 2 , K Meeran 2 & R Dina 1

1Department of Histopathology and Cytopathology, Hammersmith Hospital, Imperial College, London, UK; 2Department of Endocrinology, Charing Cross Hospital, Imperial College, London, UK.


Galectin-3, a beta-galactosidase binding lectin, has been reported to be preferentially expressed in thyroid malignancies by many authors. Moreover, it has been claimed that galectin-3 is a useful adjunct to fine-needle aspiration (FNA) in the diagnosis of follicular thyroid lesions, a notorious pitfall of this test. Although galectin-3 does seem to be expressed more often in malignant thyroid lesions, especially papillary carcinomas, it is frequently present in benign lesions and is therefore not consistent in picking up malignancy.

The aim of this study is to investigate whether galectin-3 is an accurate marker of malignant thyroid tumours, particularly follicular carcinomas, in the hope of achieving a definitive answer and to establish its possible role in the earlier diagnosis of thyroid lesions by staining cytological samples.


Retrospective study: 92 cases were chosen from the lists of thyroid histology (1997-2002) at the Hammersmith Hospital. 15 histology cases had corresponding FNA slides.

Prospective study: Thyroid samples were collected from 8 patients undergoing thyroid surgery and imprints taken. Histological samples were obtained for these cases.

All histology and cytology slides underwent immunohistochemistry using galectin-3 as the primary antibody. All slides were scored for galectin-3 expression between 0 (unstained) and 3 (strongly stained).


Retrospective study: 66.7% of the lesions had differing cytological and histological scores for galectin-3 expression.

No significant differences in galectin-3 staining were observed between follicular adenomas and follicular carcinomas.

The sensitivity and specificity of galectin-3 staining were 57.1% and 60.0% respectively.

Prospective study: All of the imprint slides were unstained for galectin-3 regardless of the histological diagnosis of the lesion.


Galectin-3 does not discriminate between follicular adenomas and carcinomas.

In addition, galectin-3 is not specific or sensitive enough to be used satisfactorily and cost effectively in clinical practice as a marker of thyroid malignancy.

Volume 7

23rd Joint Meeting of the British Endocrine Societies with the European Federation of Endocrine Societies

British Endocrine Societies 

Browse other volumes

Article tools

My recent searches

No recent searches.